close

Enter

Log in using OpenID

CAE Poster Award 2013

embedDownload
0010111010100
111
01
01
1
010100101
0
01110
1
0
10
011
0
10
1
01
10
00
Newsletter
Simulation Based Engineering & Sciences
Year
A CAE based procedure to
predict the low velocity
impact response
n°4 Winter 2013
The Fundamental Role of
Simulation-Based Approach
in New High Technology
Product Development
0110010
1101
001
01
01
11
Parametric CFD analysis of an
EMbaffle Heat Exchanger
10
Warm Hydroforming
Process Of Aluminum
Alloys Using LS-DYNA
Advanced shape for
robotic torque sensor
CAE Conference: 1000 engineers
at the annual event on simulation
Evaluation of Grinding Repair
through modeFRONTIER RSM
and ANSYS Mechanical
Formiamo i protagonisti
dell'innovazione
CONSORZIO TCN
Tecnologie per il Calcolo Numerico
CENTRO SUPERIORE
DI FORMAZIONE
Competenze sempre più elevate e aggiornate
giocano un ruolo decisivo per la competitività
delle imprese e la qualità del lavoro
Il Consorzio TCN offre attività di Alta Formazione per
ingegneri. Per il 2014 sono previsti:
• Un’ampio catalogo di corsi a calendario ricco di
proposte formative per la diffusione delle discipline
che afferiscono alla simulazione numerica
• Percorsi formativi personalizzati, costruiti sulla
base delle specifiche esigenze dell’azienda
• Formazione Continua per accrescere le proprie
competenze, nel contesto aziendale o professionale,
finanziati dai Fondi Paritetici
Interprofessionali
• Un portale dedicato alla formazione a distanza
per l'ingegneria
Alcune tematiche proposte nei corsi: meccanica
della frattura, acustica, progettazione a fatica,
tematiche legate all’attrito e usura, meccatronica,
dinamica dell’impatto, data mining, analisi
dinamica con applicazioni agli elementi finiti...
FLASH
For many of us, this time of the year is a time for reflection. We think about our values, our goals and how
we can fulfill our dreams. The same applies to our careers and businesses. Values like responsibility and
sustainability never change, they are the foundations for innovation and entrepreneurship.
While the year turns to an end, EnginSoft looks back on the recent International CAE Conference which
brought together an expert audience of about 1000 delegates from around the globe. We are all driven by
the same motivation, our belief in numerical simulation and its tremendous impact on successful product
development and research - Today and in the future.
For the 2nd time, the Conference also hosted the International CAE Poster Award, a competition that recognizes
the outstanding work with CAE technologies by teaching and research bodies and students. We are delighted
to present the awarded posters in this Newsletter.
This Edition also informs us about Selex, a Finmeccanica Company and international leader in electronic and
information technologies. Luigi Paris, head of mechanics & PCB, updates us on the importance of simulation
and human engineering skills. EMbaffle describe their parametric CFD analysis with ANSYS Workbench
and ANSYS CFX. Politecnico di Bari illustrates FE investigations with LS-DYNA for the warm hydroforming
process of aluminum alloys.
Moreover, we hear about a CAE-based procedure for a composite CAI specimen, and how the coupling of
LS-DYNA and modeFRONTIER supported the work of the Italian Aerospace Research Centre.
SACMI is an international group that manufactures machines as well as complete plants for the ceramics,
packaging, food and plastics industries. Their case study focuses on the evaluation of grinding repairing
operations. SACMI also introduces us to the company’s customized fatigue solution: ACT, the ANSYS
Customization Toolkit.
EnginSoft Nordic reports about high fidelity simulation while Mentor Graphics and
EnginSoft Italy outline the successful implementation of a Design of Experiments
approach in Flowmaster.
Further articles that the Editorial Team has collected for our readership cover the
latest achievements in robotics, in urban design and system engineering, as well as
in injection molding simulation at the company INglass.
Our Software News discuss the latest capabilities of the ClinicOptimizer by
Lionsolver, the ANSYS Mechanical Release 15, MAGMA, SW Forge, ESAComp
4.5, Scilab and Rocky, a powerful DEM package marketed by Granular Dynamics
International.
The MUSIC and WIN-shoes Projects along with [email protected], our joint venture with
IDESA S.r.l, present some of our corporate news in this last edition of the year 2013.
I would like to take this opportunity to thank our customers, readers and partners
for their loyalty, their business and the excellent knowledge exchange through the
years. It gives all of us at EnginSoft great pleasure to wish you and your families a
healthy, very happy and prosperous New Year!
Stefano Odorizzi, Editor in chief
3 - Newsletter EnginSoft Year 10 n°4
Flash
Sommario - Contents
CASE HISTORIES
6
The Fundamental Role of Simulation Approach in
New High Technology Product Development
8
Parametric CFD analysis of an EMbaffle
Heat Exchanger
11
A CAE based procedure to predict the low velocity
impact response of a composite CAI specimen
14
Finite Elements Investigations About The Warm
Hydroforming Process Of Aluminium Alloys
Using LS-DYNA
17
20
24
26
Advanced shape for robotic torque sensor
Evaluation of Grinding Repair through
modeFRONTIER RSM and ANSYS Mechanical
ACT (ANSYS Customization Toolkit)
SACMI customized fatigue solution
Accurate Thermo-Fluid Simulation in
Real Time Environments
High Fidelity Simulation of turbulent reacting flows
30
SOFTWARE UPDATE
33
36
ESAComp 4.5: new simulation capabilities for a
wider customer base
Urban design and system engineering:
risks and opportunities
Engineer your fire!
38
Rocky Discrete Element Method Package
41
TESTIMONIAL
42
INGLASS: oltre la camera calda
RESEARCH AND TECHNOLOGY TRANSFER
44
WIN-shoes: When Innovation makes Shoes
Contents
46
47
CLINIC OPTIMIZER Interactive visualization for your
personal and intelligent choice of medical treatment
MUSIC Project – First Review Meeting
CAE CONFERENCE
48
50
51
52
International CAE Conference 2013
CAE Conference 2013: will ANSYS Mechanical
Release 15 satisfy technical user expectations?
Discussion and final considerations at the ANSYS
Mechanical meeting
Scilab at the International CAE Conference 2013:
what a great session!
CAE Conference 2013 sessione MAGMA:
un grande successo
53
54
The International CAE Conference 2013 welcomed
participants from Japan
Forge NxT: l’Italian User Meeting raddoppia
CAE POSTER AWARDS
55
56
Stenting in Coronary Bifurcations: Image-Based
Structural and Hemodynamic Simulations of Real
Clinical Cases
CAE Poster Award 2013
Design by Optimization of a Controllable
Pitch Marine Propeller
CFD characterization and thrombogenicity analysis of
a prototypal polymeric aortic valve
Thermo-Fluid Dynamics model of two-phase system
58
60
62
64
alloy-air inside the shot sleeve in HPDC Process
Stenting in Coronary Bifurcations: Image-Based FEM
Analysis, Modelling and Control of a Hexacopter
EVENTS
66
WEBINAR CFD e supporto fluidodinamico
Corsi di addestramento software 2014
Perché la simulazione al computer sia davvero utile ai processi progettuali e
produttivi dell’industria, occorre potersi fidare dei risultati che essa produce.
E la correttezza dei risultati dipende sia dalle scelte fatte nella predisposizione
dei modelli - in relazione al problema che si vuole trattare ed alle caratteristiche del software utilizzato – che dalle modalità di controllo dei risultati. Non si tratta di un passaggio scontato: da un lato l’aspettativa dell’utilizzatore
dei software quanto a soluzioni semplificate sul piano formale è pressante, e dall’altro la complessità dei problemi che possono essere trattati – sia
sul piano delle dimensioni dei modelli, che delle fisiche che possono essere rappresentate – è sempre crescente. Ne consegue un contesto in cui, nel
caso di software commerciali, il ‘fai da te’, anche nel caso di utilizzatori con una buona preparazione culturale, può essere rischioso e, quanto meno,
espone a perdite di tempo ineconomiche e, a volte, frustranti.
E’ per questo che EnginSoft, da sempre, considera l’addestramento all’uso delle tecnologie software il servizio più importante da offrire ai propri
clienti, ed impiega, nei propri corsi, i migliori specialisti di cui dispone al proprio interno e nella rete dei propri consulenti. Addestrare all’uso di un
software, una volta capite le scelte relative all’architettura del sistema e all’interfaccia utente, significa prestare attenzione alle problematiche applicative. Un software ‘general purpose’ per l’analisi meccanico-strutturale permette di affrontare lo studio di un componente massivo, ma anche di valutare
la risposta sismica di un edificio, o di ottimizzare il comportamento di un elastomero in grandi deformazioni: situazioni, tutte, che richiedono modelli
specifici, e la capacità di dominare analisi specifiche, sia sotto il profilo concettuale, che sotto quello numerico. Così un software per la simulazione
dei processi che si manifestano nella colata di un metallo, richiede una precisa definizione di proprietà termo fisiche dei materiali, nella forma adatta
allo schema numerico assunto, non sempre intuibile con ragionamenti di buon senso.
Perché, allora, perdere tempo in tentativi, e rimanere con il dubbio che quanto si sta facendo non sia corretto? Molto meglio chiedere all’esperto
come utilizzare correttamente la tecnologia, e come realizzare i propri modelli, acquisendo rapidamente, e con sicurezza, le conoscenze necessarie a
lavorare bene nel proprio settore. L’offerta EnginSoft di corsi di addestramento software è molto ricca.
La si può consultare su www.enginsoft.it/formazione. Vale la pena considerare questa opportunità: essere addestrati correttamente paga!
Newsletter EnginSoft
Year 10 n°4 - Winter 2013
COMPANY INTERESTS
To receive a free copy of the next EnginSoft Newsletters, please contact our
Marketing office at: [email protected]
EnginSoft GmbH - Germany
EnginSoft UK - United Kingdom
EnginSoft France - France
EnginSoft Nordic - Sweden
www.enginsoft.com
EnginSoft S.p.A.
24126 BERGAMO c/o Parco Scientifico Tecnologico
Kilometro Rosso - Edificio A1, Via Stezzano 87
Tel. +39 035 368711 • Fax +39 0461 979215
50127 FIRENZE Via Panciatichi, 40
Tel. +39 055 4376113 • Fax +39 0461 979216
35129 PADOVA Via Giambellino, 7
Tel. +39 049 7705311 • Fax +39 0461 979217
72023 MESAGNE (BRINDISI) Via A. Murri, 2 - Z.I.
Tel. +39 0831 730194 • Fax +39 0461 979224
38123 TRENTO fraz. Mattarello - Via della Stazione, 27
Tel. +39 0461 915391 • Fax +39 0461 979201
10133 TORINO Corso Moncalieri, 223
Tel. +39 011 3473987 • Fax +39 011 3473987
www.enginsoft.it - www.enginsoft.com
e-mail: [email protected]
The EnginSoft NEWSLETTER is a quarterly magazine
published by EnginSoft SpA
Cascade Technologies www.cascadetechnologies.com
Reactive Search www.reactive-search.com
SimNumerica www.simnumerica.it
M3E Mathematical Methods and Models for Engineering www.m3eweb.it
ASSOCIATION INTERESTS
NAFEMS International www.nafems.it • www.nafems.org
TechNet Alliance www.technet-alliance.com
Advertisement
For advertising opportunities, please contact our
Marketing office at: [email protected]
RESPONSIBLE DIRECTOR
Stefano Odorizzi - [email protected]
PRINTING
Grafiche Dal Piaz - Trento
The EnginSoft Newsletter editions contain references to the following products which are trademarks or registered trademarks of their respective owners:
ANSYS, ANSYS Workbench, AUTODYN, CFX, FLUENT and any and all ANSYS, Inc. brand, product, service and feature names, logos and slogans are registered trademarks
or trademarks of ANSYS, Inc. or its subsidiaries in the United States or other countries. [ICEM CFD is a trademark used by ANSYS, Inc. under license]. (www.ansys.com)
modeFRONTIER is a trademark of ESTECO Spa (www.esteco.com)
Flowmaster is a registered trademark of Menthor Graphics in the USA (www.flowmaster.com)
MAGMASOFT is a trademark of MAGMA GmbH (www.magmasoft.de)
FORGE is a trademark of Transvalor S.A. (www.transvalor.com)
5 - Newsletter EnginSoft Year 10 n°4
Contents
Autorizzazione del Tribunale di Trento n° 1353 RS di data 2/4/2008
All pictures are protected by copyright. Any reproduction of these pictures
in any media and by any means is forbidden unless written authorization by
EnginSoft has been obtained beforehand. ©Copyright EnginSoft Newsletter.
CONSORZIO TCN www.consorziotcn.it • www.improve.it
The Fundamental Role of Simulation-Based
Approach in New High Technology Product
Development
Selex-ES is a company distinguished by its high-technology methods
and products. To be successful in this leading-edge environment,
Selex-ES must carefully define its optimal design strategy, making
best use of the available tools and approaches. In particular, the Finite
Element Method is a central component of this whole process, playing
a fundamental role from the initial scoping and costing of the project,
through its technical design and on to the evaluation
of product performance. This article considers specific
examples to demonstrate how the various stages of the
Selex-ES process are strengthened by the utilization of
such simulation methods.
Introduction
Simulation is nowadays strongly connected to new
product development in most high-technology industries;
a trend accelerated by the growth in high-performance
computing and improvements in simulation tools made
possible by innovations in computer hardware, software
and the conceptual understanding of the underlying
physics of the simulated processes. As a result, simulation
technology is now deeply-rooted in all our product development: in
fact, it constitutes an essential component of our tenders, is central
to our design process and in many cases may be embedded in our
product itself or form part of our end-user technical report.
The support of simulation in the preparation of tenders
Companies such as Selex-ES will often use simulation support during
the preparation of Technical Tenders. At this stage it is extremely
important to evaluate the relationship between technical performance
requirements and global design costs. In many cases simulation support
helps the Bid Team to correctly estimate the necessary costs in order
to satisfy the technical requirements. In extreme cases, the simulation
approach may be able to identify economic and technical reasons that
Case Histories
can produce an insight into whether the project is worth pursuing due
to high cost of delivering the stated technical requirements.
The support of simulation during preliminary design phase
A simulation-driven approach is usually fundamental to the preliminary
design phase. It is an important process in the thorough definition
of all the technical requirements to each subsystem that forms part
of the final product. Making good quality early design decisions for
each subsystem is central to the management of the design process,
enabling the early identification of issues that, if missed, could result
in corrections later in the design process: typically, the later an issue
is identified, the more expensive will be its correction. For example if
early simulation is used to correctly calculate the thermal or structural
Newsletter EnginSoft Year 10 n°4 - 6
requirements between a mechanical box and the PCB of an item of
electronic equipment intended for an avionics application, accurate
final results can be forecasted. If a methodical simulation approach
has been defined, this typically leads to better project decisions
at this stage than the alternative approach of solely relying upon
experience-driven ideas of “best practice.” The figures illustrate
the results from some preliminary Finite Element models aimed at
identifying any critical issues related to the thermal requirements
between the mechanical chassis and the PCB substrates.
The support of simulation during detailed analysis
Detailed Design is characterized by a huge use of simulation in
various different fields:
• Thermal
• Structural
• Fluid-dynamic
• Electromagnetic
The goal of all these calculations is to address the mechanical
packaging of the product; providing the designer with the necessary
guidance to:
a) Achieve the requested technical requirements.
b) Prepare for the experimental tests that will be necessary to verify
the simulation results, having in mind the reduction in cost of this
essential phase of the project – typically, experimental test will be
costly and should not be more extensive than strictly necessary.
Fig. 1 - Electronic Equipment :Thermal Map on external surface
Fig. 2 - Temperature increase of inside air
The support of simulation during the engineering test phase
At this stage, the physical properties and behavior of the test
equipment itself becomes very important and so its various
attributes (stiffness, mass etc.) must be accurately represented in
the simulation environment. For example, in a durability (shaker) test
it will be necessary to represent:
• The anchorage chassis to the shaker table.
• The presence of any air channels (assuming the product is
tested in a wind tunnel).
• The movement of the shaker table.
All these activities are carefully represented by simulation to verify
that the best (most representative) simulation and, therefore, test
results are obtained.
Conclusions
Today, the simulation approach has come a long way and it is not
possible to develop a new product without efficient calculation
support. Today’s rapidly-developing software tools are optimized
to make the best use of our rapidly-advancing computational
hardware. However, it is also necessary to have in the company a
human technical team able to manage this computational power and
ensure that it is able to contribute optimally to product knowledge
and performance at all stages of product development.
Fig. 3 - Temperature increase of Metallic part of Power Supply
Luigi Paris, Selex-ES
For more information:
Roberto Gonella, EnginSoft
[email protected]
7 - Newsletter EnginSoft Year 10 n°4
Fig. 4 - PSD - Endurance Analysis - Excitation along X Results along Y
Case Histories
Parametric CFD analysis of an EMbaffle
Heat Exchanger
Heat exchangers play an important role in the process industry, not
only for conditioning the process streams but also for attaining a
favorable heat economy. Several types are now available on the
market, but most common are the shell-and-tube heat exchangers.
They are used for many purposes, e.g., heating and cooling, (re)
boiling and condensing. Many varieties in size, shape and orientation
exist, depending on the service to which they are applied.
Over the past few decades several techniques were developed to
enhance the heat transfer coefficient and reduce operating costs.
On the tube-side, inserts or fin tubes can be used to enhance
turbulence (improving the Heat Transfer Coefficient), or simply to
extend the heat transfer surface, in order to get more compact and
efficient heat exchangers.
On the shell-side, several enhanced technologies are nowadays
available (EMbaffle but also Twisted Tube, Helix, RODbaffle, etc…),
although conventional segmental designs still dominate the market.
EMbaffle® Technology
EMbaffle® Technology was invented in Shell Global Solutions
International B.V. in 2002 and several patents have been granted
all over the world in the last decade. It is a shell-and-tube heat
exchanger technology where tubes are supported by expanded
metal baffles (Figure 1). These EM baffles are not solid plates
like common segmental baffles, but open grids made of sheets
of expanded metal in which the openings have a characteristic
diamond shape with specified tight tolerances.
Unlike segmental baffle exchangers, in an EMbaffle the flow direction
is longitudinal (Figure 2), so that dead zones are completely avoided
(reducing fouling tendency) and shell-side pressure drop is low. As
the grids are placed at relatively short separations and the tubes are
fully supported by each grid (figure at the top of the article), tube
vibration is also prevented.
Case Histories
Fig. 1 - EMbaffle heat exchanger outline
Fig. 2 - EMbaffle longitudinal flow example
In several applications tube vibration is a crucial issue, as it can
cause tube failure and therefore reduce the lifetime of the exchangers.
The combination of longitudinal flow and better tube support makes
EMbaffle the best solution every time tube vibration is governing
the design of the exchanger. Moreover, the presence of the grids
increases the turbulence and the low pressure drop allows more
compact designs, leading to an enhanced heat transfer coefficient.
Newsletter EnginSoft Year 10 n°4 - 8
EMbaffle® Technology is currently applied in several industrial
applications, such as:
• Gas-to-gas applications (Gas fields, LNG, etc.).
• On-shore and off-shore processing.
• Refining and petrochemical.
• Concentrated Solar Power (CSP) applications.
Parametric Model
In order to better understand the local behavior of the shell-side
fluid approaching the metal grids at different operating and process
conditions, a parametric 3D numerical CFD model of a representative
portion of the heat exchanger was implemented by EMbaffle in
collaboration with EnginSoft. The model was built in ANSYS Workbench
and ANSYS CFX was used as the CFD solver. A representative portion
of the whole EMbaffle bundle was selected as the minimum repeatable
geometry to be modelled (Figure 3).
Table 1 - Geometric parameters of the numerical model
Fig. 5 - Grid picture showing some of the geometric parameters
Fig. 3 - Minimum repeatable geometry extracted from the full bundle
Due to its fabrication process, the grid
shape results in a quite complex geometry
(Figure 4).
In order to simulate it in a realistic way,
several parameters related to grid shape
and tube dimensions were set in the
model. By means of other parameters it is
also possible to modify the baffle spacing
(distance between two consecutive grids)
and the number of grids. In Table 1 all the
geometrical parameters of the model are
summarized and briefly explained, while in
Figure 5 the most relevant ones are showed.
9 - Newsletter EnginSoft Year 10 n°4
Fig. 4 - Grid section
Analysis Cases
As a deeper knowledge of the local phenomena is indeed important
in order to optimize the design of the heat exchanger, specifically for
some ranges of Reynolds number or “particular fluids”, this CFD work
was planned to study the local interactions between the shell-side
medium and the grids at different operating and process conditions.
Starting in 2002, experimental tests with pilot EMbaffle heat exchangers
had been performed. Results of the tests carried out
by two internationally recognized institutions (HTRI
and TüV NEL), were used to develop the general
correlations that are nowadays applied to design
an EMbaffle heat exchanger. Some of the test cases
used by HTRI to develop the EMbaffle correlations
were then selected to guarantee a proper validation
of the CFD model.
In the present paper, a liquid-to-liquid case with
n-pentane flowing shell-side and water flowing
tube-side is examined. Two analysis cases with the
following turbulence models were set:
• Analysis Case 1: Shear Stress Transport for both
shell-side and tube-side fluids;
• Analysis Case 2: Baseline (BSL) Reynolds Stress
for the shell-side fluid and Shear Stress Transport
for the tube-side fluid.
Case Histories
Table 2 - Boundary conditions
Fig. 8- Results comparison table
Fig. 6 - 3D streamline of the shell-side fluid approaching the grid
Results
In the Figure 6 the behavior of the shell-side fluid approaching the
grids is shown.
As it may be seen, there is an increase in the fluid velocity when it
crosses the grid; however, just after the grid itself the presence of a
vortex region governs the development of the turbulence and so the
enhancement of the performance. The same phenomenon can be more
clearly observed in Figure 7, where the streamlines in a longitudinal
vertical section are shown.
The comparison between the experimental data and the two CFD
Analysis Cases is summarized in Figure 8. It is possible to notice that
the exchanged duty in the CFD cases is significantly lower than the
one retrieved from the experimental tests; the mismatch is 23.4% for
the first analysis case and 20% for the second.
Even if a small improvement is registered when using the BSL
Reynolds Stress approach, results show that the applied turbulence
models are not yet able to match the real experimental data, both for
the turbulence and for the overall heat transfer coefficient.
Fig. 7 - Streamline plot on a vertical plane
Case Histories
Conclusions and future developments
The aim of the work was to study the local interaction between the
shell-side medium and the grids at different operating and process
conditions. Results show for both analysis cases that the CFD model is
too conservative with respect to the real performance of the exchanger
used for the tests. As we are still in a validation phase of the model,
further analysis still needs to be performed.
After this, we will use the CFD model as a tool to better design the
EMbaffle exchangers for some particular applications. For example,
we could use it to investigate the optimum baffle spacing in terms of
minimum pressure drop and maximum heat transfer coefficient. More
generally, such models should permit us to address the effectiveness
of a selected grid when applied to specific fluids properties.
Francesco Perrone, Marco Brignone, Marco Rottoli - EMbaffle
For more information:
Michele Andreoli, EnginSoft
[email protected]
EMbaffle® is a world leader in innovative heat
transfer solutions that offers global industry clients
significantly improved operating efficiencies with
reduced energy consumption, emissions and
costs. By combining technical innovation with
extensive operating experience the company
provides practical solutions that are showing
tangible benefits for refining, chemical and solar
power plant units worldwide. Patented EMbaffle®
heat exchanger technology offers a step change
in shell and tube type heat exchanger design.
Expanded metal baffles (tube supports) create
an open structure allowing for longitudinal flow
at the shell side. Stagnant or ‘dead zones’ found
in traditional segmental baffle heat exchangers,
which tend to foul rapidly, are not present in
EMbaffle® heat exchangers. Tube vibration
is also eliminated due to the longitudinal flow
characteristics and pressure drop is lowered.
Newsletter EnginSoft Year 10 n°4 - 10
A CAE based procedure to predict the low velocity
impact response of a composite CAI specimen
The residual strength, in particular the compression strength after
damage due to low velocity impact, is one of the most critical issue
for composite laminates. Indeed, composite structures submitted
to low energy impacts reveal a brittle behavior and can undergo
significant damage in terms of matrix cracks, fiber breakages and
delaminations. Such damage is particularly dangerous because it
may be undetectable by visual inspection and can drastically reduce
the pristine mechanical characteristics of the structure. Generally
the behavior of composite materials with respect to this issue is
experimentally evaluated by the standard CAI (Compression After
Impact) test. For this reason, in order to simulate the impact event,
an LS-DYNA FE model of this test was developed and coupled with
modeFRONTIER. The integrated procedure allowed to obtain a better
understanding of the influence of some numerical parameters on the
simulation results (sensitivity analysis), moreover the configuration
which provided the best agreement with the experimental data
(optimization analysis) was computed.
Test Case Description
Experimental impact tests were carried out according to the ASTM
D7136 regulations to assess the capability of the procedure for
investigating the impact event. A rectangular plate (150 mm x 100
mm) was impacted at an energy level of 50 J by a hemispherical
steel impactor with a diameter of 20.0 mm and a mass of 8.64 kg.
The material of the plate was a laminate composite with a symmetric
lay-up of 28 plies [45/-45/45/-45/0/0/90/0/0/45/-45/0/90/0]s.
The plies were stacked and cured in an autoclave and the resulting
average cured plate thickness was 5.012mm. The specimen was
held on a rigid fixture with a cut-out by means of four rubber clamps.
The impact support fixture is shown in Fig.1. The contact force, the
impactor velocity and displacement were recorded during the tests.
Ultrasonic c-scans were performed after each test to measure the
damaged area.
11 - Newsletter EnginSoft Year 10 n°4
Fig. 1- Impact support fixture
Fig. 2- FE model
LS-DYNA FE model
As the plates’ length and width dimensions were large compared to
their thickness, a 2D modelling approach was chosen. In particular,
layered shell elements with an element length of 3mm were used.
LS-DYNA’s linear-elastic composite shell material model (MAT54)
was adopted, based on the failure criteria by Chang FK and Chang KY.
The separation of adjacent plies due to normal or shear loads,
referred to as delamination, absorbs impact energy and decreases
the laminate stiffness and therefore needs to be covered by the
model as well. Because delamination cannot be represented inside
the continuum shell elements, the laminate was divided into a
Case Histories
Fig. 3- Sketch of the modeFRONTIER – LS-DYNA workflow
certain number of overlayed sub-laminates connected by tiebreak
contacts which were allowed to separate during the simulation
according to a specified failure law. The influence of varying the
number of layers of shell elements with these interleaved tie-break
delamination contacts was investigated, using models with 2, 3,
7, 17, and 28 layers. A model with just 1 layer of shell elements
without delamination was also investigated.
The most realistic description of the phenomenon was provided
by a model with 17 layers: in this model, adjacent plies with a
difference in orientation lower than 90° were grouped into a unique
layer of shell elements. This was chosen for further investigations.
The impactor was modelled as a spherical rigid body with
conventional shell elements and the material model MAT_RIGID.
An initial velocity of 3.36 m/s was imposed on the impactor using
the PART_INERTIA card. A very fine mesh was adopted in order to
correctly compute the contact force between the impactor and the
plate. The FE mesh used in the model is shown in Figure 2. Finally,
an automatic surface-to-surface contact with the option SOFT=0
was defined between the composite plate and the rigid impactor.
slimt in the MAT54 card) and the degradation factor
for compression failure (variable slimc in the MAT54
card). Each time a new combination of their values
was proposed by modeFRONTIER, the LS-DYNA
input file was updated and a new LS-DYNA analysis
performed in batch mode. The output of each
simulation was then post-processed and the results
of the analysis evaluated. The outputs used in this
study were the contact force time history, the plate
deflection time history, the absorbed energy and
the damaged area size. Three of these output were
evaluated directly in LS-DYNA (contact force, plate
deflection, absorbed energy), while the damaged area
was evaluated using ANSYS FE by means of an APDL
macro. These numerical results were compared to
experimental data during the post-processing phase
and the relative errors were computed. Such errors, which will
be indicated respectively as “err_f_min”, “err_d_min”, “delta_
energy” and “min_del_area” were thus the objective functions to be
minimized. In the block labelled “DOE” (which stands for “Design of
Experiments”) the user can generate an initial population of designs,
each possessing a different combination of input variables. Starting
with the results obtained from these initial designs, the “Scheduler”
block iteratively generates completely new designs with the aim of
achieving the defined goals using various optimization algorithms.
In order to study the interaction between the input variables and
the four chosen objectives a statistical analysis was performed by
evaluating an initial population of 81 designs generated by using
the Full-Factorial method with 3 levels for each variables. The
scatter matrix chart, which is a very useful tool to analyze the data of
a statistical analysis, is shown in Figure 4a).
It was found that the variable slimt is the more significant input
variable (high correlation with the 4 objectives). All parameters
were found to affect significantly the damaged area objective. All
objectives are positively correlated, indicating that the objectives
were not conflicting.
A multi-objective optimization analysis with the algorithm MOGAII was then performed. The optimization strategy evaluated 137
designs (the initial 49 Full Factorial designs followed by 88 designs
specified by the MOGA-II algorithm), leading to several candidate
modeFRONTIER – LS-DYNA process integration
In order to better understand the influence of such parameters on the
simulation results, a sensitivity analysis was performed by coupling
the LS-DYNA FE model with modeFRONTIER, a process integration
and design optimization tool. modeFRONTIER is able to explore
the design space (i.e. the permitted values of free parameters)
and find configurations which satisfy several objective functions.
The integration of the LS-DYNA FE
model described above into the
modeFRONTIER environment is
roughly described by the workflow
in Figure 3. The blocks on the top
define the input variables for which
a suitable range of variations was
set. These input variables included:
the damping constant (variable
sf in the DAMPING_PART_MASS
card), the shear strength for tiebreak
contact (variable sfls in the tiebreak
CONTACT card), the degradation
factor for tensile failures (variable Fig. 4 - a) Scatter matrix chart; b) 4D Bubble Chart
Case Histories
Newsletter EnginSoft Year 10 n°4 - 12
optimal solutions. These can be easily detected in the 4D bubble
chart of Figure 4b, where each solution is represented by a coloured
bubble of a particular size. A good configuration which minimizes
all four objectives should therefore be blue, have a small diameter
and lie towards the bottom left of the chart. Design 189 (indicated
by the red arrow) was considered to be a good compromise in
achieving these goals.
The correlation between the numerical results obtained with this
configuration and the experimental data, in terms of damaged area
size, contact force, deflection, absorbed energy time histories and
force versus displacement trend, are shown in Figures 5, 6a, 6b,
6c and 6d, respectively. The comparison shows that the fitted
simulation results and experimental data to be well-correlated.
Conclusion
An LS-DYNA – modeFRONTIER
coupled procedure was proposed
to simulate low velocity impact on
composite plate. The procedure
allowed the study of the influence
of some numerical parameters
on the simulation results and
identified a configuration which
provided the best correlation
between the numerical results and
the experimental ones in terms of
contact force, deflection, absorbed
energy time history and damaged
area envelope.
The procedure took advantage
of modeFRONTIER’s automation
capabilities,
allowing
the
calculations to run automatically
and unattended for 24 hours each
day until completed. Once validated
on an experimental database, the
procedure will permit the study of a
range of factors (material properties,
boundary conditions, stacking
sequence etc.) on the impact
resistance of a component. Hence,
damage resistant structures can be
designed by reducing the number of
expensive experimental tests.
The Aerospace Company: CIRA
CIRA was created in 1984 to manage PRORA, the Italian
Aerospace Research Program, and uphold Italy’s leadership
in Aeronautics and Space. CIRA is a company with public and
private sector shareholders. The participation of research bodies,
local government and aeronautics and space industries sharing
a common goal has led to the creation of unique test facilities,
unmatched anywhere in the world, and of air and space flying
labs. The CIRA is located in a 180-hectar area in the immediate
vicinity of Capua, in the province of Caserta, north of Naples. Its
has a staff of 320 people, most of which are engaged in research
activity within domestic and international programs.
Fig. 5 - Correlation between the numerical and experimental results in terms of damaged area size
Rosario Borrelli, Stefania Franchitti,
Francesco Di Caprio, Umberto Mercurio
- Italian Aerospace Research Centre
Vito Primavera, Marco Perillo EnginSoft
For more information:
Vito Primavera, EnginSoft
[email protected]
13 - Newsletter EnginSoft Year 10 n°4
Fig.7 - Correlation between the numerical and experimental results in terms of a) contact force time histories; b) absorbed energy
time histories; c) deflection time histories; d) force versus displacement trend
Case Histories
Finite Elements Investigations About The Warm
Hydroforming Process Of Aluminum Alloys
Using LS-DYNA
The present work investigates the Warm
Hydro Forming (WHF) process of an
AA6xxx series alloy (AA6061-T6) using
a numerical-experimental approach.
As concerns the experimental activity,
tensile and formability tests in warm
condition were carried out to identify
the mechanical properties: flow stress
curves, Forming Limit Curves (FLCs)
Fig. 1- Experimental equipment: a=tensile, b=formability and c=press
and anisotropy values according to
temperature and orientation respect to the
rolling direction were obtained. In addition WHF tests were carried
high drawing ratio and capability to create complex shapes; however,
out using hydroforming facilities of the laboratory of Advanced
it is used for small specific productions due to high cycle times and
Forming and Manufacturing (http://afmlab.poliba.it); the following
initial economic investments. The Warm Hydroforming (WHF) uses
process parameters were investigated: maximum oil pressure
heat to increase formability of Al alloy. The higher temperature acts
(pmax), Blank Holder Force (BHFmin and BHFmax) and working
at the crystallographic level activating additional sliding planes and
temperature. As concerns the numerical activity, Finite Elements
increasing formability. The effect is remarkable even if moderate
simulations were focused on the best modeling of the WHF process:
temperatures are adopted.
models were tuned in order to fit experimental results; in particular
In the following sections the results from numerical simulations,
different values of the coefficient of friction (COF) and various
aimed at modeling the WHF process, are detailed. The FE models
yield criteria were assumed for fitting experimental data in terms of
were created using material data from preliminary experimental
thickness reduction on the formed component. The Forming Limit
tests, which allowed to determine the mechanical properties of the
Curves (FLCs) adoption, since determining the sheet formability,
investigated alloy (AA6061-T6). In addition, results from WHF tests
allowed to identify the critical areas (possible cracking or wrinkling).
allowed to tune the FE model by fitting the experimental thickness
The post-processing was made by LS-PrePost.
distribution on formed parts.
Introduction
Low density, high strength and stiffness are some of the features
that make Al alloys interesting enough to replace some mild steels
in automotive and aerospace fields. Hydroforming is an alternative
stamping process where the “punch” is replaced by a fluid under
pressure, usually oil, that has physic-chemical characteristics such
as not to degrade at high temperatures. Actually, this technique is
largely accepted by the industry for the benefits associated with it:
Case Histories
Experimental Tests
The experimental tests were carried out with two objectives: (i)
the mechanical characterization according to temperature and
orientation (with respect to the rolling direction); (ii) the WHF
process investigation in order to have a real thickness reduction
along a preferential path to be used for calibrating the coefficient
of friction for numerical analyses. The investigated blank had an
initial thickness of 0.5 mm. Tensile tests were carried out using
Newsletter EnginSoft Year 10 n°4 - 14
Fig. 2 - Material data obtained by tensile and formability tests
a standard 20 ton electromechanical INSTRON machine equipped
with the heating device shown in Figure 1a: it is composed by 9
radiant heaters positioned all around the specimen and managed by
a PID controller in order to reach and maintain the target temperature
(+/- 1%); the front opening allows the optical measurements
system ARAMIS to acquire the complete strain field using the Digital
Image Correlation (DIC) technique: in such a way the Lankford’s
coefficients of the alloy were obtained, since tested specimens were
extracted along three different orientations: the rolling one (α=0°),
the transverse one (α=90°) and the intermediate one (α=45°).
Formability tests were carried out using the
Nakajima equipment shown in Figure 1b,
assembled on the same tensile test machine. It
allows to heat the specimens by the hemispherical
punch and to acquire the strain field by the
sensors of the DIC system. WHF tests were
carried out by the 500 kN electro-hydraulic
press machine (http://www.gigant.it) shown in
Figure 1c; it is equipped with a heated die able to
reach the maximum temperature of 300°C, an oil
pressurizing unit able to work (using heated oil) at
the maximum pressure of 350bar. The information
obtained from the mechanical characterization
have been summarized in Figure 2 in terms of flow
curves (a), anisotropy values (b) and Forming
Limit Curves (c).
NUMERICAL MODELING
Model Set Up
The commercial FE Explicit code LS-DYNA
was used for numerical simulations. The blank
geometry and the die shape utilized for simulations
are shown in Figure 3.
The blank was divided into two parts: the internal
one, which is subjected to oil pressure and the
external one (in contact with the blankholder) on
which the closing force (or BlankHolder Force,
BHF) is applied. Due to the symmetry, only half
15 - Newsletter EnginSoft Year 10 n°4
of the blank was modeled in order to reduce computational
costs; in addition also a mass scaling technique was adopted.
The blankholder and the die were modeled as rigid parts
(*MAT_20 in LS-DYNA) while the blank as deformable;
different yield criteria were used to model the material
behavior; in particular, the following anisotropic yield criteria,
in plane stress condition, were taken into account: Hill
1948 (*MAT_122), Barlat 1989 (*MAT_36) and Hill 1990
(*MAT_243). Both Lankford’s parameters (R00, R45 and R90)
and plastic flow curves along the investigated orientations
(0°, 45° and 90°) were used for determining the parameters
of the adopted yielding models. Also the isotropic yield
criterion (*MAT_18) was used for comparison purposes.
In this work the following process parameters were adopted
(for both experimental and numerical tests): Temperature
(T): 110°C; BHF: from 63 (BHFmin) up to 89 kN (BHFmax) by
a linear profile; maximum pressure (pmax): 48 bar by a linear
profile. Such process parameters were implemented through the LSDYNA cards *LOAD_RIGID_BODY and *LOAD_SHELL_SET for BHF
and pmax respectively (the adoption of the working temperature was
simulated using the material behavior specific of that temperature).
Step Analysis and Post-Processing
The COF value was evaluated by minimizing difference between
thickness data along the longitudinal middle path (axis of symmetry)
of the component. In particular, an optimal value was determined for
every yield criterion investigated in the present work by comparing
thickness results from numerical simulations with correspondent
experimental data obtained using the DIC system Aramis. The
Fig. 3 - Die and sheet design
Figure 4 - Numerical and experimental thickness distributions along the symmetry path
Case Histories
Fig. 5 - Flatness values calculated using different yield criteria
Fig. 6 - Simulation results in terms of major and minor strains of all sheet
elements compared to the experimental FLC
graph in Figure 4 summarizes numerical results (in terms of
thickness profiles along the symmetry path) obtained using the
investigated yield criteria: Barlat’89 and Hill’90 (R = f (ε)) appears
to be the ones which allow the best fitting of experimental data. In
particular, the Hill’90 criterion is able to fit better the left part of
the experimental curve (characterized by smaller strain levels) while
the Barlat’89 criterion allows to fit better the right part of the curve
(the one concerning the deepest part of the component). Also an
additional parameter was investigated for checking the robustness of
numerical models: the Flatness (it is as the ratio between the length,
LC, of the symmetry path in contact with the die and the length, LD,
of the bottom part of the die). The Figure 5 shows the flatness values
calculated using models adopting different yield criteria: using as
reference the experimental value of 0.2141, the Barlat’89 allowed
the best approximation.
In order to predict critical areas characterized by an elevated risk of
ruptures or wrinkling, material FLCs (which represent the limit values
of major and minor strains) were implemented in the numerical
models. The experimental FLCs shown in Figure 2 were used in LSPrePost as reference for the principal strain values calculated in the
FE analyses for all the sheet elements. The Figure 6 shows results
obtained using anisotropic Barlat89 model with COF equal to 0.068.
It is possible to note that none sheet element exceed FLCs curve,
Case Histories
therefore risks of rupture are not highlighted. The quality of
formed component is quite good, recording a severe thinning in
correspondence of deepest part of the component. The map of the
quality areas is corroborated by that of the thinning.
Conclusions
This work shows the LS-DYNA capability to simulate the WHF
process and the importance of the extensive alloy mechanical
characterization. The implementation of the anisotropic yield criteria
as Hill’90 or Barlat’89 seems to be the best way to fit experimental
data as the thickness reduction. It is important to underline that
it is necessary to determine an appropriate COF in order to fit
experimental thickness data. The management of the experimental
FLCs in LS-PrePost provides an easy way to show the results and to
identify dangerous areas.
V.Piglionico, G.Palumbo, A.Piccininni, P.Guglielmi - Politecnico di Bari
A.Taurisano - EnginSoft
For more information:
Antonio Taurisano, EnginSoft
[email protected]
Newsletter EnginSoft Year 10 n°4 - 16
Advanced shape for robotic torque sensor
New development in robotics has required the use of flexible
joints. The success or the failure of that structures, heavily
influences the robot motion cause of the interaction with the
control system. Nowadays the control force is undergoing a
rapid development and torque sensors represent a crucial
measurement system for it. The solution presented, thus, is a
new optimized torque sensor design that not only fits perfectly
with the current mechanism assembly, but also it guarantees
the required mechanical properties of the joint. The use of the
CAE technique allows the possibility of test several solutions
before reaching the final one.
Introduction
In recent years, robotics is moving away from the rigid joint to
look forward the design of fully sensorized joints. Generally,
in a force control loop, these ones allow the motor to tune
the torque applied to the end effector. There are several
reasons which have prompted this improvement. The
application of flexible joints:
1. make safer the hand to hand collaboration between
robots and humans;
2. allows the storage of the energy due to an impact for
avoiding structure high damages;
3. increases the precision in a manufacturing process.
Figure 1 - (a) HyQ robot, (b) Torque sensor position
Figure 2 - Flexible joint schema
The object of our work is a flexible joint for the HyQ (the Hydraulicallypowered Quadruped designed in the Italian Institute of Technology),
shown in Fig. 1. HyQ weighs about 80 kg, is 1 m long and 1 m
tall with fully stretched legs. That platform is designed to perform
high dynamic task like jumping, running, climbing, etc. The actual
version, of the robot, is able to perform both indoor than outdoor
17 - Newsletter EnginSoft Year 10 n°4
operations like: walking up to 2m/s, jumping up to 0.5m, balancing
the ground disturbance. In the future that robot can help man in
several dangerous situations like earthquake, fire, etc.
The fully sensorized joint (Fig. 2b) schema is shown in Fig. 2. It will
be installed between the motor (Fig. 2a) and the harmonic drive (Fig.
2c). This last component will address the torque to the HyQ leg.
Case Histories
Joint torque requirements
The deformation could be measured in different ways as follows:
electrically, based on electromagnetic phenomena, digital
processing or optically. In this case we used the electrical one
with strain gages; they are glued on the specific position on the
structure that deflects under the applied torque. This deflection
must be within the material elastic range to avoid hysteretic
phenomena. If the yield point of the material is overpassed, in fact,
it will be impossible to get the true value of the torque transmitted
by the motor: most of the load applied will be absorbed by the
material plastic deformation and the stress-strain relation will be
non-linear.
In this case, the flexible structure is a 1 degree of freedom (DOF)
torque sensor whose deformations are estimated by positioning
strain gauges. The design boundary conditions, as happen for
each measurement tool, depend on the admissible stiffness k,
(k=torque/deg). That parameter can influence the reaction time of
the control system: the decrement of the stiffness induces a loss in
the system accuracy. In the case presented, the maximum dynamic
torque transmitted by the motor was around 140Nm and the related
angle is around 1 deg.
The strain gauges measurement technique is based on the
principle of the local deformation. It means that it is important to
concentrate the maximum strain of the body in a specific area,
where the sensors will be positioned. This value, lower than 0.06%,
produces a suitable input for the strain gauge within the linearity
that is 0.15%.
The aforementioned surface must be accessible and planar. The
accessibility is useful to guarantee the correct positioning of the
sensor: it needs the cleaning of the surface, the bonding of the
film and sufficient space for the cables. The planarity, instead, is
important to avoid offset and drift phenomena that can arise in case
of curved surfaces.
The movements of the leg can be both in the clockwise direction
than in count clockwise, for this reason it should be better having
the same calibration factor in both directions. That will be an easy
integration of the structure in the control system.
Results
The final shape was obtained by an optimization process divided
in four case studies. Every single case study was composed by
two different steps: the initial shape was obtained according
to the applied design rule. Then it was optimized thanks to
modeFRONTIER simulations, Fig. 2, in order to determine the best
performances. This procedure was looped till these performances
matched the torque sensor requirements.
Figure 4 - Constraints
As it is obvious, the torsion is the basic phenomenon of torque
sensor concept. That leads to think that the circular shape can
produce the maximum deformation in safe conditions. Thus
the first idea would have been to design a hollow cylinder to
be installed between the harmonic drive and the shaft motor.
However, following this way, it was not possible to match the
requested deformation. Then other designs were investigated
and all the structural capabilities of each solution were tested by
numeric simulation developed in Workbench. The specifications
imposed to use Ergal as building material. The element chosen for
the mesh was solid186, because it guarantees the best refinement.
Two different constraint conditions were tested, to simulate the
reciprocal displacement between the two sides. In case of Fig. 2a,
the torque sensor is fixed to the harmonic drive and the torque
is applied by the shaft motor. The case of Fig. 2b represents the
inverse.
Based on the aforementioned criteria, the first solution investigated
is shown in Fig. 4a. The maximum stress, arose close to the
Figure 3 - modeFRONTIER schema
Case Histories
Figure 5a - 1st solution investigated
Newsletter EnginSoft Year 10 n°4 - 18
The shape of the third solution has followed the torsional flow
stress. The structure was more robust than in the previous cases
to guarantee the elastic response of the body. The deformable
components, the two horizontal links in Fig 4c, were subjected
to a pure tensile stress. The strain was bigger than the previous
times and allowed the reaching of the design specifications.
The robustness has allowed to remain in the elastic field of the
material. However, even if the positioning strain gauges surfaces
were planar, they were inaccessible for film bondage and the cable
working space.
Figure 5b - 2nd solution investigated
Figure 5c - 3rd solution investigated
The last solution investigated, grew up from the failures of the
previous one. The robustness and the idea to get the deformation
thanks to a pure tensile stress were saved. The main work was
focused in the increase the empty space close to the planar surface
to guarantee a successful strain gauges position. That solution
covered almost everyone the required tasks. The exception was
represented by the bi-directionality. To solve that problem, the
suggested solution was to use two different calibration factors
according to the clockwise or anti-clockwise movement.
Conclusions
The application of the CAE technique to the design process of the
torque sensor allowed the reduction of time and costs. In fact,
the physical prototype was machined only when the numerical
simulation results has fitted the design requirements.
M.D’Imperio, F.Cannella, J. Goldsmith, C. Semini and D.G. Caldwell
Figure 5d - 4th solution investigated
connection between the inner circular section and the linear
beam, exceeded the yield point of the material. The area where
the maximum strain was located was not planar and the reached
value of deformation was not sufficient for the strain gauges
measurement. However that shape guaranteed a bidirectional
behaviour. Considering this solution did not match several
requested specifications, it was abandoned.
The second structure has been designed more robust then the
former one, to have stresses reduction. Despite that improving,
some problems of the previous solution still existed. The only one
solved was the value of deformation, it was within the measurement
range of the strain gauges. At the end, even that second solution
was deleted.
19 - Newsletter EnginSoft Year 10 n°4
Figure 6 - Physical torque sensor
Case Histories
Evaluation of Grinding Repair through
modeFRONTIER RSM and ANSYS Mechanical
SACMI is an international group manufacturing machines and
complete plants for the Ceramics, Beverage & Packaging, Processing
and Plastics industries. This world-wide group comprises about 70
companies.
The case study under investigation is about the evaluation of the Effect
of Grinding Repairing Operation through a user friendly tool based on
modeFRONTIER’s Response Surface Methodology (RSM) capabilities.
The grinding operation is one of the methods used by SACMI to repair
the surface of its cast iron structural components when affected by
unacceptable defects like porosities or inclusions. The grinding
repair alters the component geometry, the stress field and the fatigue
parameters of the studied component. The usual way to evaluate the
effect of the grinding operation in terms of grinded component fatigue
life is a new simulation followed by a fatigue analysis where the model
geometry has to be modified according to the grinding operation. The
aim of the present study is the creation of a user friendly and fast
analytical Excel Worksheet to replace the modeling, simulation and
fatigue analysis of the ground component, utilizing modeFRONTIER
and its Response Surface Model capabilities.
Two different parametric models have been built in order to consider
two different geometric topologies in the same project and to fill the
whole grinding tool field of application. The CAD models and the
related FE models have been integrated into modeFRONTIER directly
with two Workbench nodes. The fatigue analysis follows SACMI inhouse procedures and rules and is implemented in an Excel node at
the end of the project flow. The final outputs are given to the users in
terms of Safety Factors and come from the integration between three
different RSMs obtained with modeFRONTIER and some analytical
calculations.
Having been validated, this tool is currently used in SACMI to evaluate
the effect of the grinding repair operation and allows a significant
saving of time and cost.
Case Histories
Introduction
Cast iron components are affected by different kinds of superficial
defects connected to the casting process, such as slag inclusions,
porosities and shrinkages. Depending on their type, dimension and
position, such defects may not be acceptable in terms of component
wear resistance. In these cases the component has to be repaired. One
repair option is grinding the defect to remove it. A standard flowchart
to evaluate the acceptable of such a grinding repair consists of an
FEM and fatigue analysis of the re-ground component. This approach
is accurate but time consuming: it can be used to verify the repair
and not to manage it. A different approach is proposed here: it is
based on a “grinding metamodel” created with a modeFRONTIER
RSM. Thanks to this approach the flowchart reduces to one simple
and fast step utilizing a user-friendly tool in the form of an excel GUI.
The analysis time is reduced from hours to minutes. modeFRONTIER
is used to manage the RSM, ANSYS Mechanical to compute FEM
Fig. 1 - Standard and proposed Workflow
Newsletter EnginSoft Year 10 n°4 - 20
analysis and a Microsoft Excel
algorithm focused on cast iron
SACMI structural components
to compute fatigue analysis.
The main aim of this study is to
obtain a robust and user-friendly
tool to enhance productivity of
the SACMI Imola S.C. Ceramic
Engineering Department.
Fig. 2 - Geometric topologies of the dig
The fundamental steps to obtain
the proposed tool are investigated in the next sections and include
the creation of an appropriate RSM metamodels, the construction of
a user-friendly evaluation tool and validation with some test cases.
Metamodel based on Response Surfaces
A metamodel based on Response Surfaces (RSM) is an analytic
model that approximate the multivariate input/output behavior
of complex systems, based on a limited set of computationallyexpensive simulations. RSM was first introduced by Box and Wilson
in 1951, who suggested the use of a first-degree polynomial model
for approximating a response variable. The sequence to create a
metamodel in engineering can be summarized in four steps. Model
formulation: identifying the problem’s input and output parameters;
after the creation of the metamodel
(i.e. surface treatments, surface
roughness, probabilistic factors and
the presence of lubrication). The
outputs of the metamodel should
then be two response surfaces: one
for the stress intensification factor
and one for the volume effect. To
obtain these, some preliminary
aspects should be analyzed.
Geometry of the dig. The geometry of the dig influences both the
RSM outputs. Two different topologies of dig were considered and
represented in Figure 2: one obtained from a spheroidal cutting tool
with the center outside from the original external surface (1) and the
other with a conical cutting tool with rounded tip (2). This choice
comes from the awareness that the only two manageable and relevant
dig dimensions are the radius at the bottom of the dig, R, which
corresponds to the grinding tool radius, and the dig’s depth, b. These
two are two input parameters for the RSM.
Stress field. The model should take into account all the possible
stress fields. Considering the typical depths of grinding digs, a
linearized stress field is a good approximation. Under this assumption
the generic stress field is obtained by the linear superposition of a
purely normal load and a purely bending load. Once the reference axial
Fig. 3 - ANSYS WB project and ANSYS Mechanical environments
Design selection: using a DOE tool to specify the variable settings at
which to run the disciplinary models and acquire response data. RSM
fitting: having chosen a particular type of RSM, its parameters are
adjusted to best match the data obtained during design selection. RSM
Assessment: specifying and evaluating the performance measures that
will be used to characterize the fidelity of the fitted RSM. The RSM
can be then used for various purposes such as the prediction of the
responses of unevaluated designs, optimization, trade-off studies or
the further exploration of the design space. modeFRONTIER contains
powerful tools to create, manage and export Response Surfaces
Models.
Concepts behind the metamodel
The metamodel’s aim is to evaluate the consequences for component
resistance of geometry changes introduced by a grinding repair. The
geometry of a grinding repair influences two different fundamental
aspects of the component fatigue analysis: the stress field and the
volume effect. It is known that both these aspects strongly affect
the resistance of a cast iron component. The other factors which
influence component resistance are not connected to the grinding
repair geometry so they are introduced directly in the Excel tool
21 - Newsletter EnginSoft Year 10 n°4
load is fixed, the generic load is completely defined by the bending
moment (i.e. the bending moment parameter is the exponent of the
applied moment magnitude, a, so M=1∙10a). The response surface of
the stress intensification factor is a linear combination of two different
RSMs: one obtained from the pure axial load and the other from the
pure bending load.
Mono-axial load.The FEM model is characterized by a mono-axial
load even if a general load field could be multi-axial. This assumption
has been evaluated by a preliminary sensitivity analysis with multi-axial
loads. Without this simplification of the FEM model, the metamodel
would become hard to manage in terms of user inputs required.
Scalar intensification factor. The use of a scalar intensification
factor to obtain the new stress field with the dig starting from the
original one implies that the principal vector directions do not change
among the two geometric situations.
ANSYS Workbench project
The parametric ANSYS WB project is made of three ANSYS Mechanical
static structural environments. They all share the material, geometry
and model data. The parametric geometry has been created in Design
Modeler. The first static structural environment is a linear static
Case Histories
Fig. 4 - modeFRONTIER workflow
analysis with the reference pure axial load and symmetry boundary
conditions. The second is a linear static analysis with the pure bending
load with magnitude defined by the input parameter a and symmetry
boundary conditions. The third is a dummy static analysis for volume
effect evaluation during fatigue analysis.
modeFRONTIER RSM
Defining a correct initial DOE (Design of Experiments) is fundamental
to obtaining a good Response Surface. Since the model is not very
expensive from a computational point of view while the inputs ranges are
quite large, a large population has been used (almost 300 design points).
The purpose was to comprehensively cover the full design
modeFRONTIER Workflow
space especially in the most common grinding repair
The whole process to create the
ranges. Two different DOE algorithm have been chosen:
the Incremental Space Filler algorithm was used in the first
Response Surfaces of the grinding
repair has been completely automated
stage, starting from few design points of SOBOL, to cover
the full design space and then in the second stage the
through modeFRONTIER. In particular it
effectively evaluates fatigue analysis of
SOBOL algorithm only was used to increase the number
the repaired component model through
of design points in some specific domain areas where a
better resolution of the RSM was required. Figure 5 shows
the following steps:
the DOE design points in a scatter 3D plot. After running
the DOE sequence the design points were split in two,
1. Choose the geometric topology
through a switch node.
forming training and validation sets. The first is used to
2. Launch an ANSYS WB project to
create the RSM and the second to validate it: a good RSM
Fig. 5 - DOE design points, 3D scatter plot
perform static structural analysis.
(trained on the training set) should be a good predictor of
3. Launch an Excel Worksheet with
the validation set. The high number of design suggests the
macro to perform fatigue analysis.
selection of a suitable RSM amongst approximating (not interpolating)
4. Export the relevant outputs.
surfaces. Figure 6 shows the 3D surface graphs of the chosen response
surfaces. They are obtained from Multivariate Polynomial Interpolation
based on the Singular Value Decomposition (SVD) algorithm. A 5th
Figure 4 shows the modeFRONTIER Workflow. On the left, the two
degree polynomial was used for kt_axial and kt_flex responses while
possible routes (1 or 2) that the flow can take are highlighted,
a 4th degree for Volume Effect. Using modeFRONTIER RSM evaluation
depending on the geometric input values, b and R. On the right, the
tools one can get an average relative error less than 1% and a maximum
inputs (blue) and the outputs (brown) of the metamodel are circled
relative error of 10% for all the three surfaces. Using the Chauvenet
and the Excel node is squared in green.
criterion few outliers were found and neglected during the RSM creation.
Fig. 6 - Response Surfaces (Design points in black)
Case Histories
Newsletter EnginSoft Year 10 n°4 - 22
Fig. 7 - Excel evaluation tool GUI
Grinding evaluation tool
The RSM metamodel described above was then packaged in a userfriendly tool. It runs in Microsoft Excel since this is a well-known
and widely available item of software which incorporates its own
GUI creation environment. Furthermore, modeFRONTIER offers a
direct export of its response surfaces to Excel. The tool GUI as it
appears to the user and the evaluation tool workflow are summarized
in Figure 7 and Figure 8 respectively. The tool workflow is briefly
summarized here.
The user puts the main inputs values: geometric inputs (b and R)
and original stress field inputs in terms of Safety Factors without
considering the volume effect, [email protected] The Safety Factor, SF, is
one of main outputs of a fatigue analysis, it is proportional to the ratio
between the reference stress limit, σlim and the equivalent stress, σe:
SF proportional σlim/σe. The quantities σlim and σe depend on the
fatigue approach followed during the analysis. In particular the one
used here is an in-house multi-axial approach focused on typical
hydraulic press structural components. The tool gets the information
of the original stress field through the equivalent stress which contains
the information of the full stress tensor. In particular the user input
is the Safety Factor which is inversely proportional to the equivalent
stress since it is directly available among the classical fatigue outputs.
To make the tool more user-friendly, the user need only supply one
scalar stress value and not a complete stress tensor. However, this
implies the following approximation: the stress gradients are assumed
to be the same before and after grinding repair. In particular the tool
Fig. 8 - Excel evaluation tool workflow
23 - Newsletter EnginSoft Year 10 n°4
needs the SF on the surface, [email protected]_sup, and at b depth, SF@
Vref_b, at the defect position.
The tool evaluates the stress field gradient (parameter a) from SF@
Vref_sup, [email protected]_b and b. The tool evaluates the volume effect
through the Volume Effect RS from b and R. The tool evaluates the
stress intensification factor through kt_axial and kt_flex RS from a,
b and R. The user provides the additional input parameters which
influence component resistance (material, surface treatments, surface
roughness, probabilistic factors and the presence of lubrication)
for which the tool utilises its internal database to compute the
consequences. The relative dimensional changes during grinding are
the only information required by the tool since it reads SF and returns
SF. The tool evaluates and returns the minimum SF combining the
information from the stress intensification factor, volume effect and
the additional inputs.
Evaluation tool validation
The complete tool has been validated with a comparison between its
results and the ones obtained with the standard grinding evaluation
workflow for a set of available test cases. As shown in Figure 9 all the
tested cases give results within a ±5% band of the reference results.
This reveals a high confidence level for the completed tool.
Conclusions
A new approach and tool to evaluate grinding repair has been obtained
thanks to RSM. The modeFRONTIER software appears to be a very
efficient tool for the creation and management of such RSMs, and
the integration of at least three different software packages (Design
Modeler, ANSYS Mechanical and Excel) and two different geometric
topologies. Deriving the tool requires some effort to produce the
modeFRONTIER workflow, execute the analyses and process the
results. However, once the RSMs are derived and incorporated within
the tool, the analysis is very much faster than the previous approach,
taking minutes rather than hours. Furthermore, the approximations
that are used have been shown not to generate unacceptable errors,
with validation demonstrating robust results.
Riccardo Cenni, SACMI Imola
For more information:
Francesco Franchini, EnginSoft
[email protected]
Fig. 9 - Validation results
Case Histories
ACT (ANSYS Customization Toolkit)
SACMI customized fatigue solution
ACT overview
ACT is a completely new and fully documented customization
environment available since version 14.5 of ANSYS Workbench.
It allows legacy ANSYS MAPDL expertise to be reused in order
to produce straight-forward customized solutions for the ANSYS
Workbench Mechanical environment. The developed customized
items can easily be used by any CAE analyst without the need to
possess MAPDL legacy expertise.
SACMI customized fatigue assessment
Having to deal with about 20,000 tons of Ductile Cast Iron (DCI)
per year, SACMI has developed its own methodology for the
fatigue assessment of this material. The core of the methodology
is a probabilistic, multiaxial and volume dependent local stress
approach. The inputs needed, for each node are the stress tensors
at the two relevant stages of the working cycle and the associated
volume. The outputs are probability of failure, safety factor and
extension and the severity of some typical DCI defects. The latter
can be used either in the design or in the Quality Assurance phase.
ACT implementation
In order to have the SACMI fatigue assessment implemented as a
user-friendly tool inside ANSYS Workbench, EnginSoft developed
an ACT extension able to guide the user from the initial settings
down to the results visualization.
The procedure is dived into three different steps accessed from a
custom toolbar developed using the ACT customization framework.
• Exporting the data for Excel: from the first button on the
toolbar, a customized post-processing item is added to
the ANSYS Mechanical module tree. It lets the user define
relevant parameters for the fatigue assessment such as the
surface finishes throughout the component and the time steps
to be considered. During solution, an automated procedure is
Case Histories
started in order to export stress tensors, volume and surface
finish at each node in csv format.
• Performing the fatigue calculations and exporting data
to WB: a proprietary Excel file developed by SACMI is opened
by the second button on the dedicated toolbar. It reads csv
files previously generated and allows the user to set up other
relevant model parameters and to then perform the fatigue
assessment. As a result, a set of outputs are generated for
each node, from the safety factor to the allowable residual
stress state. These data are then exported in a fashion suitable
to be imported back into ANSYS Workbench.
Newsletter EnginSoft Year 10 n°4 - 24
• Visualizing the results in ANSYS Workbench: ACT takes
care also of the last part of the process via the third and final
button on the toolbar, which automatically imports the fatigue
results and displays them on the original model, using the
External Data tool available in ANSYS Workbench. The user
therefore has access to all the usual features to probe the
results, slice the model and so forth.
Conclusions
Thanks to the specific customization capabilities available with the
ACT framework, it has not only been possible to completely embed
a complex procedure inside the ANSYS Workbench environment,
but to also give it many added values, including:
• Units automatically managed by ANSYS Workbench.
• Relevant geometry selections managed via the standard
ANSYS Workbench features.
• Everything “behind” the procedure hidden from the end-user,
making the capabilities available to a wider set of users who
need not be aware of the details of the methodology.
• The solution should not be vulnerable to future updates of
ANSYS.
Matteo Cova, SACMI
For more information:
Francesco Micchetti, EnginSoft
[email protected]
SACMI is an international group
manufacturing machines and complete
plants for the Ceramics, Packaging (including
Beverage and Closures&Containers), Food
and Plastics industries - markets in which
it is a recognized worldwide leader. Its
strength lies in the application of innovative
technology, the outstanding position of
the Group on international markets and its
commitment to research and development
and providing customers with top-flight
quality and service.
25 - Newsletter EnginSoft Year 10 n°4
Case Histories
Accurate Thermo-Fluid Simulation in
Real Time Environments
The need for comprehensive and repeatable system-level testing
of embedded systems can present major economic and technical
challenges to systems and test engineers. It often involves combining
real hardware components with a software simulation model to perform
hardware-in-the-loop (HIL) simulation and testing. This technique is
essential in evaluating and verifying systems that cannot easily or safely
be tested in a real operating environment and where testing extends to
failure mode analysis of the system.
The HIL simulation concept has many applications from a relatively
simple AC temperature controller to a more complex system such as an
aircraft re-fuelling system. HIL simulation requires real time interaction
with the software simulation model that represents part of the system
environment under test.
The progress in numerical simulation methods and high-performance
computing provides thermo-fluid system engineers with greater power
to gain insights into their systems’ performance through the exploration
of multiple system configurations. In many cases, complex simulations
are not able to run natively in real time, which makes the software
solution unsuitable for coupling with a HIL environment. Therefore an
alternative approach is needed.
This paper describes the Design of Experiments approach used in the
Mentor Graphics 1D thermo-fluid simulation software Flowmaster V7
to address the issue providing simulation results in real-time. Mentor
Graphics has collaborated with EnginSoft in order to implement the
creation of Response Surface models within the Flowmaster GUI. This
framework allows for the creation of meta-models of a full simulation
model to be exported as C code or as MATLAB™ S-Functions suitable
for use in a HIL environment or as the backend code to a runtime model
of the system. The latter when implemented as a portable runtime
version with a dashboard interface provides a useful tool allowing
non-experts to review the results of a simulation model analysis and
understand a system’s behavior more easily.
Meta-Models and the Design of Experiments Approach
Using Flowmaster V7, engineers can employ a Design of Experiments
technique to perform a series of simulations that can be used to create
a response surface that interpolates all intermediate points. This
surface represents a model of the original model; in other words, it is
a meta-model that can be used to analyze the global problem over a
defined range of input conditions. The meta-models are created from
parametric studies in which one or more input variables are varied in
combination to determine the effects on selected output parameters.
The advantage of using the Design of Experiments technique is that
fewer calculations need to be performed to produce a well-distributed
set of simulation results. The Fig. 1 illustrate the C Code generation and
real time integration workflow in Flowmaster V7.
Fig 1 - Real-time integration workflow in Flowmaster V7
Case Histories
Newsletter EnginSoft Year 10 n°4 - 26
Constructing a Meta Model
Constructing a useful meta-model starting from a reduced
number of simulations is not a trivial task. Mathematical and
physical soundness, computational costs, and prediction
errors are not the only points to take into account when
developing meta-models. When using meta-models,
engineers should always keep in mind that this instrument
allows a faster analysis than the complex engineering models,
but interpolation and extrapolation introduce a new element
of error that must be managed carefully.
These are the steps to using meta-models for engineering
design, starting from a validated system model:
• First, formulate the problem and identify the problem’s
parameters; this may include specifying the names and
bounds of the variables that will be part of the design.
• If the original simulation is computationally intensive
and the use of the meta-model is necessary, choose
the number and type of designs for which it is more
convenient to run the original simulation model.
• Use output responses to build meta-models.
• Validate the meta-models.
Fig 2 - Flowmaster V7 Response Surface with Hardys Radial Basis Function
• Finally, use response prediction for determining new
conditions.
(RBFs) are used in the software because of the high tractability for
The number of input parameters times the number of simulations
these models. Moreover, empirical evidence shows that such models
forms the table of the training points used to construct the metagive good predictions even with the reduced number of training points.
models. If the training points are not carefully chosen, the fitted model
RBFs are simply linear combination of radial functions centered at
can be poor and influence the final results.
experimental points xi:
Generating Inputs for the Meta-Model
In Flowmaster V7, a “Latin Square” method is used to generate unique
combinations of distributed points in the domain.
Latin Square is a Design of Experiments algorithm based on Latin
Several different radial functions Φ are available in the literature;
squares, mathematical objects first investigated by Euler starting from
however, we opted for a subset of functions: Gaussians, Duchon’s
1782, in which numbers are never repeated in columns and rows. The
number of generated designs is n2, where n is the number of different
Polyharmonic Splines, Hardy’s MultiQuadrics, and Inverse
levels that we want to consider for the input. This approach produces a
MultiQuadrics (IMQ). This list guarantees a good degree of freedom
balanced list of experiments, where all points should be run the same
for interpolating several different problems.
number of times for each one of the n levels.
The final number of required simulations is thus independent from
Evaluating and Assessing the Meta-Model
the number of input variables; this represents an enormous advantage
Assessing the meta-model involves evaluating the performance of the
compared to other factorial methods generating regular grids. In fact,
models, as well as the choice of an appropriate validation strategy.
when n is the number of levels for the m input, full factorial requests
Validation is a fundamental part of the modeling process. Engineers
n to the power of m points and this number grows exponentially
may use residual charts and other statistical information at their
accordingly to the number of variables. For example, with three
disposal for evaluating the accuracy of the meta-models. This is
variables and five levels, 25 points are required using the Latin Square
necessary to understand the behavior of the model, improve it when
method, compared with 125 points using a full factorial approach.
necessary by adding additional simulations, or redefine the region of
Creating this kind of experiment allows engineers to populate the input
interest.
data grid in the software with sets of unique values over a specified
range simply by entering the lower and upper bounds for each defined
The maximum absolute error may be used as a measure to provide
parameter.
information about extreme performances of the model. The mean
absolute error that is the sum of the absolute errors divided by number
Fitting the Meta-Model
of data points may even be used; it is measured in the same units
The output responses from running the simulations with these inputs
as the original data. RBFs fit exactly the training points so we need
are used to fit a meta-model. Meta-model fitting involves specifying
a smart approach to check the goodness of the model. The error is
the type and functional form of the meta-model and then saving,
estimated with the “leave one out” technique in which one point is
evaluating, and comparing different responses. Radial basis functions
left out of the training and kept as a measure of the error. In turn,
27 - Newsletter EnginSoft Year 10 n°4
Case Histories
each point of the experimental set is excluded from the interpolation
and used to evaluate the residuals. This approach provides a good
estimate of the global error.
Exporting the Meta-Model
The meta-model is used to predict responses at untried inputs. Having
derived the meta-model that defines the system response, the last
step is to export it to C code or MATLAB™ S-Functions. These forms
are both suitable for use in a real-time environment. The C code also
can be used to derive a runtime model with a dashboard interface.
The dashboard interface would provide the facility of being able to
enter input values within the range of the model and immediately see
the output results allowing sharing and use of the model results with
non-experts.
Application Example
The main application areas for hardware-in-the-loop simulation are in
the design of Electronic Control Units where the controller is connected
to a real time simulator. This provides a way of testing control systems
over the full range of operating conditions including failure modes
both cost effectively and safely. Application examples come from a
range of industries such as Automotive, where HIL real time modelling
is used in the design and evaluation of electronic control systems
optimised for hybrid and electric vehicle applications and in the
design of vehicle AC and cooling systems. It is also applicable to the
Aerospace industry for the design of aircraft refuelling systems.
The example network shown below in Fig 3 is a simplified automotive
engine cooling system, where the engine is represented by a heat
source transmitted into the cooling system via a thermal bridge
component. The flow passes through a heat exchanger and there is
a bypass line controlled by a set of globe valves that represent the
thermostat.
The primary circuit consists of a pump, a heat source, a set of
globe valves, a cross-flow heat exchanger and a pressure source
Fig 3 - Simple Automotive Cooling System
Case Histories
that pressurizes the system. The bypass line includes a globe valve,
component C4, which controls the amount of fluid that passes through
the heat exchanger by modulating between position 0 and 1. If the
valve is in position 0, then a quantity of flow will pass through the
bypass line and if the valve is in position 1, then the entire flow passes
through the heat exchanger.
In this study, we want to characterize the cooling system network
performance for a range of pump speeds, air flows over the radiator
and engine heat outputs. We also want to include the effects of various
valve positions for the bypass line from fully closed to fully open.
Fig 4 - Experiment input values
The following four input parameters are defined for the network.
• [Pump Speed] Mixed Flow Pump, C16
• [Air Flow] Flow Source, C14
• [Engine Heat Output] Heat Flow Source, C1
• [Valve_C4] Valve Opening, C4
The output parameters are defined as:
• Top Hose Temperature (Thermal Bridge C2, Node 2)
• Pump Flow Rate (Mixed Flow Pump, C16)
In Flowmaster V792, we can generate the inputs to the required
simulations using a Latin Square algorithm. This method provides a
good distribution of results values within the domain that are suitable
for creating a bounded response surface model.
The normal procedure would be to use Latin Square to generate input
values based on the bounds entered for each of the defined input
parameters. However, as the model is to include valve states, from
fully open to fully closed, any corresponding single response surface
Fig 5 - Response Surface View
Newsletter EnginSoft Year 10 n°4 - 28
would show large errors due to interpolation across the open/closed
valve boundary conditions. To overcome this Flowmaster V7 allows
discrete values to be used for any input parameter, in this case the
valve, and combines the Latin Square values with each of the discrete
values. A response surface can then be generated using any two inputs
parameters, an output parameter for each discrete value of valve
opening.
Fig 6 Deviations of Response Surface for the flow rate through the pump
provided which will automatically determine and display the best fit
RBF. The result of applying a RBF to the simulation results is shown
in Fig. 5.
The Deviation Details tool provides an immediate and simple evaluation
of the goodness-of-fit of each response surface on the basis of its
deviation. As shown in Fig 6 and Fig 7 the best response surfaces for
the flow through the pump and through the heat exchanger are those
computed with Gaussian RBF while the best response
surface for the temperature in the primary circuit is the
one computed with Hardy’s MultiQuadrics.
Each combination of results for the defined inputs, each
output at a defined valve position and a selected radial
basis function can be saved as a meta-model. The metamodels can be exported as C code or as MATLAB™
S-Functions either of which is suitable for use in a
real-time simulation or as the backend to a dashboard
interface.
Conclusion
The collaboration between EnginSoft and Mentor
Graphics has resulted in the implementation of a
Design of Experiments approach to response surface
modelling in Flowmaster V7. It provides for the creation
of meta-models within the Flowmaster GUI based on
Latin Square Experiments that can be exported as C
code or as MATLAB™ S-Functions suitable as the
backend code to a runtime model of the system or for
use in a HIL environment.
The ability to characterize a system’s behaviour in
exported code opens up a wide range of possibilities,
such as creating a simple dashboard that allows
non-expert users to understand and predict system
performances, inserting the code into a hardwarein-the-loop logic, or embedding the code into other
codes for co-simulations.
Silvia Poles, Alberto Deponti - EnginSoft
Frank Rhodes - Mentor Graphics
Fig 7 Deviations of Response Surface for temperature downstream of heat source
Here a Latin Square of ten levels is considered which generates 100
simulations i.e. n2 simulations, where n is the number of different
levels. Using four discrete values for the valve component C4 of 0,
0.3, 0.6 and 1 will produce a total of 400 steady state simulations.
For this model the following values are used.
Once the 400 simulations are completed, response surfaces for each
output variable can be created using the following radial basis functions:
1. Gaussian
2. Duchon’s Polyharmonic Splines
3. Hardy’s MultiQuadrics
4. Inverse MultiQuadrics
This list of radial functions guarantees a good degree of freedom for
interpolating several different types of problems. An option of ‘All’ is
29 - Newsletter EnginSoft Year 10 n°4
For more information:
Alberto Deponti, EnginSoft
[email protected]
Reduce the development time and costs
of thermo-fluid systems
Flowmaster is the leading general purpose 1D Computational
Fluid Dynamics (CFD) solution for modeling and analysis of
fluid mechanics and pipe flow in complex systems early in the
development process. It helps systems engineers to simulate
pressure surge, temperature and fluid flow rates system-wide and
to understand how design alterations, component size, selection
and operating conditions will affect the overall fluid system
performance accurately and quickly.
Flowmaster is supported in Italy and South Europe by EnginSoft.
Case Histories
High Fidelity Simulation of turbulent reacting flows
What are High Fidelity Simulations?
High fidelity simulation (HFS) involves the use of very large
computing capabilities to resolve directly most of the physical
phenomena under consideration, hence reducing the impact
of the modelling in the results. For fluid flow simulations, it
implies resolving a large fraction of the turbulent kinetic energy
spectrum in order to limit the eventual uncertainty contained in
traditional models. In other words, we use well-resolved Large
Eddy Simulation and capture the large dominant turbulent vortices.
When chemical reactions occur in the turbulent flow, high fidelity
simulations are required to deal with a complex (at least not oversimplified) reaction scheme (typically at least 20 species for
simpler systems) covering the relevant range of time scales.
Why High Fidelity Simulations?
The growing availability of high performance computing (HPC)
opens avenues for significant changes and improvements in
Computer Aided Engineering practice within industry. While socalled production CFD is an established tool with a short leadtime, high fidelity simulation (Large Eddy Simulation based) has
previously been outsourced to universities or research institutions.
This distinction is no longer valid, since HPC and HFS are use
during the industrial design process. The challenge is therefore
to integrate judiciously high fidelity simulation into the work
flow as a complement to existing tools. One may identify three
occasions where high fidelity simulations is of value for the product
development process. Firstly, it is able to generate databases for
the validation and calibration of simpler (sometimes steady-state)
models that will be used in production CFD. Secondly, HFS is
used for trouble-shooting investigations where the production
CFD tools have failed to prevent malfunction or have overlooked
important parts of the physics. Thirdly, HFS is of value for getting
a more advanced understanding of the fluid system with the aim
of improving operations or identifying limits in terms of critical
operating points.
Case Histories
When to do High Fidelity Simulations?
Based on the points developed in the previous subsection, the
product development process benefits from HFS when dealing with a
novel design or operating conditions that lie outside the established
region of confidence of simpler tools. It is, in fact, a central tool in
innovation-driven tasks where traditional compromises are to be
challenged – hence also challenging traditional simulation tools.
In addition, HFS is a valuable tool in the later development phase
for validating the design before the production of a prototype – in
other word, during the preventative trouble-shooting phase. HFS is
well-suited for use prior to expensive experimental validations, test
campaigns or costly prototype manufacturing.
How to do High Fidelity Simulations?
A key issue in HFS is the high resolution required to capture
accurately the physics under consideration. It necessitates a fine
numerical mesh and high order numerical solutions (no numerical
diffusion, for example). The mesh quality itself ought also to be high
(preferably cubic cells) in order to make the very best use of this
high spatial resolution. Matching the spatial resolution, it is also
necessary to utilize high order time integration techniques along
with small time steps (Courant number below 0.3). An estimate
of the computational resources required will depend on the model
size and in particular the smallest / fastest physical scale to be
resolved. More importantly, unlike production CFD, HFS places
strong demands on expertise of the engineers and project manager
in charge.
Simulation of a piloted premixed jet burner (PPJB)
Firstly, we exemplify the use of HFS through the simulation
of a piloted methane jet flame. It consists of a round methane/
air premixed jet (equivalence ratio 0.5) issuing in vitiated gas
(temperature 1500K) at a bulk velocity of 50 m/s. The jet of diameter
D is surrounded by a stoichiometric pilot (temperature 2336K)
securing a stable flame thanks to the very hot gases contacting
Newsletter EnginSoft Year 10 n°4 - 30
Fig. 1- Instantaneous temperature, OH mass fraction and CO mass fraction fields for the PPJB. The premixed jet (equivalence ratio 0.5) is injected in the center surrounded by the
stoichiometric pilot
the premixed jet. It contains all the physics involved in pilot
stabilization in industrial burner and was well characterized using
state of the art laser diagnostics at Sandia National Laboratories
and the University of Sydney. It is therefore a suitable but severe
test for advanced simulation strategies. HFS was performed using
a 21-species and 84-reactions skeletal mechanism describing
methane combustion.
Figure 1 presents three longitudinal cuts depicting different
scalars. The temperature field shows clearly the location of the
pilot and penetration of the fresh mixture jet. The effect of the pilot,
in terms of promoting reactions, is also seen in the OH field with a
thin but intense layer of OH in the shear-layer. Further downstream,
the effect of the pilot fades and the reaction layer is intermittent
with alternating thin band and large pockets of CO. The statistics
of CO are therefore a good indicator of the flame dynamics and are
examined in Figure 2. The peak value is located in the shear-layer
where the jet contacts the pilot and the vitiated gas. The peaks
merge at the tip of the jet – about x/D=25. As expected, the RMS
value exhibits a large peak (up to 50% of the mean) in the shearlayer. HFS captures accurately the peak CO value both in term of
mean and RMS – with differences of the order of the experimental
uncertainty. It indicates that the dynamics of the reaction layer
are simulated both qualitatively and quantitatively. Discrepancies
are only seen close to the pilot and are in fact due to the inflow
boundary conditions of the pilot. In fact, the lack of detailed
experimental characterization of the inflow boundary conditions is
presently the larger source of uncertainty and limits the predictions
with HFS. A conclusion is that HFS is a powerful technique where
modeling assumptions have a lower impact on the results than
user expertise or uncertainties over the boundary conditions.
Simulation of ozone assisted exhaust gas cleaning
with detailed chemistry
A second example focuses on cold plasma treatment of exhaust
gases for NOx removal. The apparatus resembles the experiments
of Stamate et al. and is presented on Figure 3. It features a set
of coaxial cylinders used as reactor shells with the exhaust gas
injected on the left hand side. The purpose is the conversion of
NOx molecules (mostly NO and NO2) to N2O5 by reaction with
ozone (O3) generated by a cold plasma. The complex chemistry
depicts the oxidation of NO into NO2, of NO2 into NO3 and also the
formation of N2O5 from NO and NO2. In fact, the complex chemistry
is described by 13 species and encompasses 31 reactions.
HFS is used presently to resolve the large coherent structures in the
counter-flow problem. Both the ozone and exhaust gas jets exhibit
irregular and large scale patterns – although they are statistically
axi-symmetric (on average). Figure 3 also presents the mass
fraction fields of key species. Whereas O3 presents sharp gradients,
Fig. 2 - Time averaged and RMS carbon monoxide profiles at different axial locations (x denotes the axial coordinate) – comparisons between high fidelity LES and experimental data
31 - Newsletter EnginSoft Year 10 n°4
Case Histories
Figure 3 -Instantaneous fields in the gas treatment reactor; left: visualization of the exhaust gas and ozone jets core; right: mass fraction fields for O3, NO2 and N2O5 in a longitudinal cut
NO2 and N2O5 have smooth gradients. This arises from the variety of
chemical time scales under consideration, with very active species
such as O3, being consumed in a small volume around the nozzle,
while slower reactions proceed in the whole reactor volume. HFS
offers the possibility of capturing both time scales, as well as the
complex, multi-scale interaction with turbulence.
High Fidelity Simulations today in industry
High fidelity simulation is already an engineering tool as it has
been integrated into the product development chain. Suitable
applications typically have a very high physical test cost which
advocates – and makes attractive - massive parallel computing as
an alternative. Historically, the first sector to use HFS has been
the gas turbine and aero-engine industry. Here, combustion
chambers are subject to instabilities arising from coupling
between inherently unsteady swirling flows, flame dynamics and
the thermo-acoustics of the chamber. For military applications,
after-burners are also sensitive components that are potentially
unstable. For these two applications, only HFS is able to capture
the underlying mechanisms and give engineers the knowledge
required to improve the systems.
Besides combustion modelling, aero-acoustics is also best
captured by HFS. Two common applications are the noise created
by a jet engine exhaust and, more recently, by vehicles. Another
field where HFS is of value is the study and design of liquid
atomization – for example, for fuel injectors in piston engines or
gas turbines. No doubt that it is only a start and that several other
industrial sectors will follow.
Conclusion
High fidelity simulation is a powerful tool for handling very accurately
non-linear and chaotic systems such as turbulent reacting flows.
Thanks to the increasing availability of computational resources,
it is now mature for integration in the product development
process with spectacular achievements as illustrated above. It
gives rise, however, to some important new questions regarding
data management and system definition / knowledge. Very large
volumes of data (terabytes) can easily be generated, which must
be managed and interrogated, and a high degree of knowledge
Case Histories
of the system boundary conditions and physics is demanded. At
present, there is no accepted best-practice for answering these
two questions and the success of HFS lies in the expertise and
experience of the user.
Acknowledgments
The author thanks Dr. Matthew J. Dunn, University of Sydney, for
making the experimental data available and for fruitful discussions
on limitations and meaningfulness of comparisons between
experimental and HFS data.
For more information:
Christophe Duwig, EnginSoft Nordic
[email protected]
Image at the top of the article: separate sources of Fluorescein (green) and
Rhodamine (red) are injected on the axis of a water turbulent jet (blue),
in its downstream far field. Images from the Paper: “The mixing of distant
sources” by Mihkel Kree, Jerome Duplat - Aix Marseille University and
Emmanuel Villermaux - CEA/UJF-Grenoble
Figure 4 - HFS of a swirling flame: visualization of the flame surface colored by the
streamwise velocity, ref. Iudiciani /et al/ 2011 /J. Phys.: Conf. Ser./ *318* 092007
Newsletter EnginSoft Year 10 n°4 - 32
Rocky Discrete Element Method Package
The Discrete Element Method (DEM) is a relatively new technique which is
gaining great popularity with the advancements of computer technology.
This approach is used for the simulation of granular materials, which
consist of a large number of solid particles. Continuum equations for this
type of material are very difficult to derive for a general flow case. To avoid
this problem, the Discrete Element Method relies on the simulation of the
motion of every solid particle in the system of interest. The interaction of
granular particles with each other and system boundaries are traced at
every time step of the simulation.
Rocky is very powerful DEM package marketed by Granular Dynamics
International, LLC. It is a shared-memory parallel software which allows the
fast solution of granular mechanics problems. It has several capabilities
that are unique in the commercial DEM world; these capabilities include
true non round particle shapes, the ability to simulate breakage without
loss of mass and volume, the simulation of shape change for boundary
surfaces due to wear, amongst others. The package is extremely popular
in the mining industry and is gaining popularity for other applications
related to solid particles flows.
Brief Description of Discrete Element Method
The Discrete Element Method deals with simulations of the flow of
granular materials, consisting of many solid particles. Examples of these
material types include sand, ore, grain and so forth. These materials
are very common in engineering applications and the ability to predict
their flow characteristics is an extremely important task. However, unlike
deformable solids and fluids, attempts to derive accurate equations of
flow and motion in continuum form failed. These equations have been
found for only two extremes – the first one is static situations (the elasticplastic or rigid-plastic approach in soil mechanics) and rapid granular
flow (this is a mathematical abstraction which is not applicable for particle
flow with realistic energy dissipation and under the influence of gravity).
Unfortunately most of flow regimes for granular materials lie between
these two extremes and accurate continuum solutions for them are not
available. The Discrete Element Method is relatively new technique which
deals with this problem by “brute force” - namely by simulating every
particle of the granular material in the flow subject to contact and external
33 - Newsletter EnginSoft Year 10 n°4
Figure 1. A picture of a conveyor transfer chute simulated by Rocky DEM package and
examples of particle shapes available in Rocky
forces. With this approach one does not need to know the equations of
state and motion of granular media; only contact interaction laws are
needed and a variety of reliable contact models exist for this purpose.
Apart from that the other important advantage of DEM compared to
continuum approach is that information is obtained on the particle scale.
Sometimes this particle-scale information is essential: for example, the
prediction of particles breakage when energy applied to every particle in
the system has to be calculated.
While the idea behind DEM is extremely simple, its implementation is
not straightforward. The technique relies strongly on computer power and
efficient modern parallel programming techniques; without them a DEM
program will run for very long time and will be impractical for engineering
applications. Recent advances on these fronts have made DEM a good
practical tool for engineering simulations.
Software update
History of Rocky DEM Package
Rocky is a relatively new DEM package: the development of the code
started less than four years ago. However the code is based on the
success of in-house DEM solutions developed by Conveyor Dynamics,
Inc. from 1995. These in-house codes did not have any user interfaces;
therefore the interface for Rocky is relatively new but the solver is very
mature. The first version of the code released in summer 2011 was
designed for the simulations of transfer chutes only; later, the code
interface was updated to add grinding mill simulations. Starting from
Rocky 2.0 released in the middle of 2012 the code has been developed
and marketed as a general-purpose DEM code. The code is now being
developed by Granular Dynamics International, LLC in collaboration with
the Engineering Simulation and Scientific Software Company (ESSS).
Figure 2. Simulation of particles flow inside grinding mill. Over a million of particles were used
in this simulation
Unique Capabilities and Example of Applications
There are number of DEM codes in the market these days and a user
now has a choice of DEM packages - both commercial and opensource. Compared to these packages, Rocky has several capabilities
that are unique in both the commercial and open-source world. We are
going to describe here only the most important ones; descriptions that
will necessarily be brief in view of the space available in this paper.
First of all, Rocky was developed with actual practical engineers
in mind. The information a user will obtain from the software is not
Figure 3. Prediction of wear of a grinding mill lifters inside Rocky. Presented on the left-hand
just a collection of pretty pictures and movies but parameters that are side of the picture is slice of the mill with new lifters and on the right-hand side the same slice
important for the engineers. These parameters are power draw on all at the end of wear simulation process
moving bodies, shear and impact wear parameters, forces, flow rates
and so on. The models that are incorporate with the software are realworld physical ones: they are extensively tested both internally by the
company on many consulting projects and through our collaboration
with universities worldwide. We believe the ability to predict the realworld rather than virtual-world result is the most important characteristic
of the software.
The other important feature of Rocky package is the ability to simulate
true non-round particles. Other DEM codes rely on clusters of spheres
for this purpose, but in Rocky the shape you see on the screen is the
actual shape being simulated. This allows us to simulate shapes that
are closer to reality and also properly simulate breakage of the particles
(which is another unique feature of Rocky) without the loss of mass
or volume that is unavoidable with spherical clusters. Some examples
Figure 4. Simulation of particles breakage inside concept of a new grinding device
of particle shapes that could be created and simulated in Rocky are (CAHM -Conjugate Anvil and Hammer Mill).
presented on Figure 1. Also shown in this figure is very typical example
Breakage simulation is another important feature of Rocky due to be
of a Rocky application, simulating the performance of a transfer chute. A
released in the next version. The breakage model in Rocky combines
transfer chute is a gravity device very widely used in mining industry for
models from the mining and gaming industries for the prediction of
sharp changes of the direction of material conveyance.
particle energy, strength and fragment generation during the breakage
Figure 2 presents another example from the mining industry, the
event. An example of the model application is presented on Figure 4 simulation of a full grinding mill. This particular simulation has over a
this is a new conceptual device (Conjugated Anvil and Hammer Mill)
million particles and over seven hundred thousand boundary elements,
being developed by Conveyor Dynamics, Inc. for particle comminution
which is considered to be quite large by DEM standards. Rocky is an
processes in the mining industry.
efficient shared memory parallel code and can handle this simulation
Starting from Rocky 2.2.0 the software can be coupled with the ANSYS
quite quickly.
Structural and ANSYS Fluent packages. The coupling is one-way at this
Figure 3 presents one more unique feature of Rocky, the ability to simulate
stage, with work now in progress to provide two-way coupling in the near
physical wear of the boundaries. The software collects shear work applied
future. The forces applied by particles to the boundaries can be exported
by particles to the boundary and removes boundary volume proportional
into the ANSYS structural package and the resulting deformations can be
to this wear work. This feature is extremely useful for predicting the
calculated. An example of this type of simulation is presented on Figure 5.
characteristics of particle flows where they are affected by boundary
For this case, a simulation of the motion of particles on a vibrating screen
changes due to wear.
Software update
Newsletter EnginSoft Year 10 n°4 - 34
first coupling approach is validated against experimental data obtained at
TUNRA laboratories (University of Newcastle, Australia) for airflow around
a transfer chute and the agreement with experimental data was excellent.
The parent company responsible for the development of Rocky is in the
mining industry, with a natural consequence that the majority of its users
have also been in this sector. However during recent months the software
has been gaining popularity in other industries such as agriculture (for
example, the corn flow simulation of Figure 6), pharmaceuticals (where
Rocky is being used for tablet coating simulation), materials handling, the
construction industry (see Figures 7 and 8 – the simulation of soil flow
around a conveyor frame and the simulation of a truck loading station)
and many others. There is really no limit to the range of industries to
which Rocky may be applied - it suitable for any case where the motion
of many solid particles has to be accurately predicted.
Figure 5. Simulation of particles flow
on vibrating screen in Rocky (top)
and screen deformations simulated in
ANSYS Structural package (bottom);
the nodal forces predicted by Rocky
DEM were exported into ANSYS
was carried out and the nodal forces applied by
the particles were exported to ANSYS Structural
to permit the screen frame deformation to be
calculated, in addition to the results obtained
from Rocky package alone (such as screening
efficiency and screen wear characteristics).
The coupling with ANSYS Fluent can be done
for both the particles driving the flow of fluid
(such as the airflow created around transfer
chutes frame caused by falling ore particles)
and the fluid driving the flow of particles. In the
first case, the continuum parameters of particle
flow are calculated inside Rocky on the Fluent
mesh and provided via User-Defined Functions
to the Fluent solver. In the second case, Fluent
case and data files are read directly into Rocky
and forces applied by the fluid to the particles
are calculated inside the DEM package. The
Conclusions
Presented in this paper is a very brief description
of the Rocky Discrete Element package. This
package is an extremely powerful tool for the
simulations of the flow of granular materials.
The package can be very useful for engineers
and researchers from a variety of industries.
Figure 6. Simulation of corn flow with Rocky DEM package
Figure 7. Simulation of soil flow around shifting conveyor frame
35 - Newsletter EnginSoft Year 10 n°4
Future Plants for Software Development
Rocky has historically benefitted from very rapid development. However,
even this pace is about to see a very significant increase! Granular
Dynamics International, LLC is joining forces with the Engineering
Simulation and Scientific Software Company to develop and marked
the software. The new version of Rocky 3.0.0 due to be released early
next year will feature an advanced new interface with many new features
available for the analysis of particles flows. The two-way coupling with
ANSYS software has also been planned for the near future. We are also
working actively on the improvement of Rocky’s
speed to enable even larger problems to be
handled, tracking many millions of non-round
particles in a reasonable amount of computer
time.
Alexander V. Potapov
Granular Dynamics International, LLC
Figure 7. Simulation of soil flow around shifting conveyor frame
Software update
ESAComp 4.5: new
simulation capabilities
for a wider customer base
The new ESAComp 4.5 version, to be released at the
end of this year, proves again to be an effective and
cross field tool for preliminary analysis of composite
materials, suitable to a wider and wider customer base.
One of continuously updated features is the ESAComp
material Data Bank: more than 50 fiber-reinforced
material systems have been added, of which most
include information related to the mechanical behavior in
different environmental conditions as well as statistical
data. Furthermore, exploiting the experimental data Fig. 1 - Modal analysis on a panel reinforced on both faces
provided by the major material suppliers, the effects due
to physical properties’ statistical distributions can be
evaluated through ESAComp probabilistic analysis tool.
The other upgrades concern numerical analyses: during
the setup phase the user can see a model preview in
order to control the current configuration; after simulation
he can save the results for panel and cylindrical shell
analyses keeping the solution for later reference without
having to redo it. Furthermore an easier comparison is
possible among the behaviors of the same structure
under different load conditions, changing lay-ups or
varying stiffener configurations, with the possibility to Fig. 2 - Integration between ComposicaD and ESAComp for the winding process simulation
see the results through a tree-view. Beam stiffeners of
different types and dimensions can be combined and placed on
an efficient and economical way for reliable process in terms of
both sides of the panel/cylinder at the same time, which increases
design simulation, verification and manufacturing. The design
the versatility of the modules.
study can be realized in ComposicaD™ for the winding process
simulation, which is able to predict and map the fiber directions
A completely innovative feature is the analysis of vessels:
Componeering has integrated ESAComp tightly into the design
on the vessel’s ends, then the candidate design is exported to
ESAComp, where the FE analysis is performed. The transferred files
process of composite pressure vessels combining the dedicated
include the FE model and laminate layups for the different sections
features developed in some numerical environments, as
ComposicaD™ and ACP; this new ESAComp application provides
of the vessel, while material properties for FEA are introduced in
Software update
Newsletter EnginSoft Year 10 n°4 - 36
ESAComp. Once the solution is run, it is possible to realize very
detailed post processing and design verification.
The growing awareness and interest on analysis tools available in
ESAComp is confirmed by the consistent presence of EnginSoft
and Componeering in the most relevant international technical
events: it is clearly proved that the design and simulation tools
are nowadays essential for the industrial competitiveness.
The new features and upgrades available in ESAComp 4.5 are
developed to meet more and more users’ requirements coming
from any market level and to make the tool intuitive and accurate
at the same time.
Fabio Rossetti, EnginSoft
André Mönicke - Componeering Inc.
Fig. 3 - Post processing of the composite pressure vessel
Through-the-thickness evaluation of IRF
For more information:
Fabio Rossetti, EnginSoft
[email protected]
Fig. 6 - Geometry preview available during the analysis setup
Fig. 4 - Post processing of a composite pressure vessel – Inner Strain
www.esacomp.com
The software ESAComp
Fig. 5 - Structural analysis on a cylinder with 2 types of stiffeners
37 - Newsletter EnginSoft Year 10 n°4
ESAComp is a software for analysis and
design of composites. Its scope ranges from
conceptual and preliminary design of layered
composite structures to analyses of details.
ESAComp is a stand-alone software tool, but
thanks to its ability to interface with widely used
finite element software packages, ESAComp
fits seamlessly into the design process.
The comprehensive material database of
ESAComp forms the basis for design studies.
ESAComp has a vast set of analysis capabilities
for solid/sandwich laminates and for
micromechanical analyses. It further includes
analysis tools for structural elements: flat and
curved panels, stiffened panels, beams and
columns, bonded and mechanical joints.
Software update
Urban design and system engineering:
risks and opportunities
One could say that culture or human civilization started with cities:
populations adopted spontaneously organization and processes,
developed technologies to build cities where basically exchanging
goods, practices and ideas and offering protection were the ground
for economic and social development. During this historical process,
and particularly in recent years, the urban fabrics, already marked by
the complexity of all human organizations, became more and more
technically complex. Mankind recently developed, beyond Cartesian
thinking, specific disciplines to approach the more technically
complex systems that were to be designed and manufactured such
as airplanes, spacecraft etc. The new paradigm, that any elements
of a system can’t be designed properly if someone loses the link
to the system itself and even to the exterior of the system, looks
particularly relevant in the field of urban design. Indeed the more
advanced technologies have been introduced in urban systems
in a sectorial approach losing then the necessary transverse
approaches and generating unwanted side effects. Therefore
systems engineering and architectures looks very promising but
unlike industrial systems one shall not forget that urban systems
exist whether designed or not as highly dynamic systems operated
by existing or future populations. Evolutions are taking places even
in their boundaries (cities consume lands of adjacent territories)
and in their functions (a shipbuilding port may evolve into a new
technological centre of production). Two key principles should then
be ensured by stakeholders wishing to develop new tools adapting
industrial systems design to urban systems design: the first is to
organize the accessibility for all stakeholders, the urban political
governance and the urban technicians of course but also for the
populations, residents or not and the second being to develop
observation systems to monitor progress and provide understanding
tools of the evolving complexities to help stakeholder adjusting their
policies. A third one should be added, more technical, but is mainly
the necessary technological conclusion of the first two: the toolbox
Software update
needed must be open i.e. interoperable in between all sub-systems
and solutions, and ever evolving as cities are not built for a certain
duration but possibly for eternity!
Advancity
The subject of sustainable cities, at least in the perspective that is
used in the French competitiveness cluster called Advancity can be
structured in the following 3 axes.
The ultimate goal is to provide what we propose to call urban quality
of life: it is made of attractivity, competitivity, safety, security,
resilience, access to culture, to health, to information, to services
and to mobility. This quality of urban life provides in turn a whole
range of services to individuals and to business community to enjoy
and perform their activities. This first axis is made mostly of service
activities that take an ever larger share of the economic wealth of
cities.
Then, and in order to support this goal, cities have to develop a full
range of physical facilities and structures, very often organized in
sub-systems, that must be environmentally efficient: lean resources,
autonomy, less pollution, circular economy, CO2 neutrality, energy
consumption, limited vulnerability etc. are a few of the criteria that
must fulfill all the facilities such as public space, green and blue
infrastructure, built environment, transport systems, street networks,
houses, apartment blocks. Most of the economic value and expenses
of the urban systems lie in this line of actions.
The third and last axis is the one of governance: it covers the whole
range of activities from intelligence (all information to appreciate and
understand the urban functioning) to designing and managing cities
and achieving resilience. These activities are in the order of 10% of
all other activities only but are extremely important as they have a
Newsletter EnginSoft Year 10 n°4 - 38
with population, is key for the system of
governance to take decisions. Designing cities
systems requires in fact all of them, together
and simultaneously addressed. Urban design
is a set of management activities making sure
that cities provide quality of life in the sense
I used before: economical wealth, creativity,
services, support, health etc. Under the strict
constraint of environmental efficiency. We
have then a system to consider and design
activities to structure it: for sure, this is a
topics for systems engineering.
very high leverage effect on the cities global wealth. These activities
are at the core of our today’s topics.
This perspective being a conceptual view to structure actions, it
is not adequate for organizing the works for sub-communities of
stakeholders. We then structure the stakeholders in four groups:
ecotechnologies (basic technology providers, water, energy, wastes
sectors etc.) supply basic hardware solutions to the communities of
ecoconstruction (buildings and infrastructures) and of ecomobility
(transport systems, services to travellers). These two communities
construct and operate the two main physical infrastructures of
cities and the services thereof, themselves offering their system
solutions to be integrated into a global sustainable city governed by
municipalities with the assistance of architects and engineers. IST
has a special place, transverse to all others. Information technologies
are in no way specific to cities or to construction but they irrigate
all segments and communities. One could even argue that they
are becoming – in their basic function of treating and exchanging
information – the new city engine on top of the mobility engine which
was so far the only possibility to exchange goods and to encounter
other individuals. But this would bring us too far and I
won’t elaborate further.
Urban design activity
Having so structured cities and stakeholders, let us
turn towards the urban design activity. What is it made
of? Without trying to be exhaustive, one may mention:
drafting master plans, designing public spaces
whereby urban planners and architects draw 2D plans
to visualize cities spatial structure and activities.
Another activity is the administrative task of granting
permissions to individuals to use land in a certain
way predetermined by master plans. Designing cities
means also constructing streets, large monuments
and buildings, and collective transport systems. It
is also supporting education activities, organizing
social life and ensuring economical wealth. Last but
not least, one can mention that having dialogues
39 - Newsletter EnginSoft Year 10 n°4
It is then just natural to consider cities as
complex systems: within a given territory
having certain boundaries – a large question
Fig. 1
in itself – cities are a system (or a system of
systems) made of human beings, interacting
firstly in between themselves and with technical systems providing
them with resources and services, so that they may perform their own
activities. But cities interact also with the outside world. This picture
is a very simplistic representation but everybody having some
knowledge of systems architecture can understand and imagine how
systems architecture can help in developing a proper description of
such a complex system (Fig. 1).
If we now turn into some basic facts and figures, we can size issues
and opportunities. Oil reservoir modelling is a very important activity
for oil and gas fields’ developments. Reservoir engineers are an
important segment of the key human resources being employed by
oil exploration companies. I then thought that making comparisons
between oil fields and cities may have some relevance to this
question (Fig. 2).
At a rate of 100$ a barrel, and a cost of 2 to 3 000 €/m2 for the
built environment, one can see that the asset value of a city of 1
million inhabitants is well above that one of an “elephant” oil field
and that cost and revenues are an order of magnitude greater in the
case of cities as in the case of oil fields. If you consider that oil fields
Fig. 2
Software update
have a duration life of 20 years when cities view
themselves as established for hundreds of years
if not for eternity, we can see that we do not
need to trigger a big improvement in efficiency
of cities to justify even a small investment in
design and modelling activities and tools. There
can be several reasons to justify the fact that
cities modelling developments are yet limited.
Among them, there are good arguments from
a technical nature that support the fact that the
technical challenges are enormous (Fig. 3).
The first thing to notice is that urban systems
or cities did not wait for systems engineering to
exist and develop. Cities concentrate activities
of human beings since Neolithic age and did
become more and more complex over time (in both human sciences
and “hard” sciences approaches). They are extremely evolutionary by
nature, in both their functions and pattern of activities (a commercial
city can evolve into an industrial one, or an administrative city:
look at the various facets of Rome, Berlin, Paris, Rotterdam, Bonn,
Xi’An to name just a few), and in their boundaries: cities consume
permanently their hinterland and one can’t now determine if there
is a “legitimate” boundary between Rome and Ostie, or Dortmund
and Köln. And these evolutions take place further to inhabitants and
businesses taking decisions to come or to emigrate and not just to
decisions by Mayors or elected representatives.
If now someone wants to apply systems engineering, he will have to
develop a model – virtual representation of the reality with several
simplifications – describing in a manner as close as possible to
reality, not necessarily the whole city system but probably more
likely a restricted view of it. He will use these models for planning and
managing. In order to be successful, the main operational condition
seems to me to be that the model system must still be operated
under the control of informed people and business stakeholders. If
this wouldn’t be the case, then we would see reality and models
evolving separately in different directions: something like what
would happen when driving learner wants to go left and the driving
teacher wants to go right. The reverse engineering would have to be
a recurring activity!
To sum up, the key points of attention should be:
• Reverse engineering.
• Accessibility and transparency.
• Time is very long term.
• Models must be agile and evolutionary.
• Multidiscipline, multireference, multi-models must be native:
for instance the built environment on the ground is referenced
by the cadastre, but the underground environment does not
bear any reference at all to cadastre; economics interact with
population fertility and with ground water pollution and wastes
production etc.
• Zooming in and out are entry tools in the models necessary to
consider all scales.
Software update
Fig. 3
• Interoperability from stakeholders down to software and
databases is a must.
• Data models and meta models are probably what is needed first
to structure the works of stakeholders and what can be more
generic from one model to another, from one place to another.
• Education of professionals and of population must go all along
the technical developments.
In order to be successful in facing the challenges that I sum up again:
• Time is infinite.
• Boundaries are subjective and issue dependent.
• Singularities will evolve.
• Phenomena are multiscales and multidisciplines (2D cadastre
and GIS vs 3D geomodelling…, populations, économy)
• Cities are archetypes of complexities.
• Interoperability in all dimensions.
I am of the opinion that the key success factors should be in adopting,
at least, the following attitudes and lines of actions:
• This can only be a collective venture: urban professionals; IT
population and business stakeholders; research IT/hardware/
urban/social.
• The size of the issue, and the very collective nature of cities,
requires collective funds raising tools.
• The knowledge to be developed is basically a collective property
at global and meta scales.
• Business models should be a mix between open source and
proprietary developments.
• Deliverables should address simultaneously tools for education
and for professionals.
A possible route is that of the “open source” consortium, like GOCAD
in reservoir engineering, even though GOCAD is not exactly open
source, but just to take it as an example of how good are the tools
that a truly collaborative attitude can deliver.
I hope this is a route business and research stakeholders in urban
matters and in IST may take, sooner rather than later.
Vincent Cousin, Advancity
Newsletter EnginSoft Year 10 n°4 - 40
Engineer your fire!
In occasione della International CAE Conference 2013 è stata
presentata [email protected], una joint venture tra EnginSoft S.p.A. e
IDESA S.r.l che si propone di operare nell’ambito della Fire Protection Engineering secondo logiche proprie della concurrent engineering.
[email protected] intende andare oltre la semplice collaborazione tra
due imprese e, per questo motivo, a partire da gennaio 2014 avrà
lo status giuridico di rete di imprese: un’azienda di fatto che sarà volta all’ideazione, allo sviluppo, alla
realizzazione ed alla commercializzazione di soluzioni verticalizzate in settori applicativi fortemente
legati alla sicurezza al fuoco.
La moderna Fire Protection Engineering (FPE) è di per se basata
su concetti di progettazione integrata e mira a coordinare tutti i sistemi di protezione al fuoco (attivi
e passivi, incluso lo studio delle
vie e delle modalità di esodo) in
una strategia generale di lotta al
fuoco ma è generalmente praticata applicando in modo prescrittivo
quanto previsto dalle normative
di riferimento, un approccio che
non garantisce una uguale efficacia (rapporto costi/benefici) nel
passare da un caso industriale
all’altro. Questo limite può esse-
41 - Newsletter EnginSoft Year 10 n°4
re superato adottando il cosiddetto approccio performance-based ed [email protected], individuando il proprio payoff nell’acronimo
F.I.R.E. (Fire Integrated Revolutionary Engineering), intende massimizzarne i benefici proponendosi (in virtù delle proprie elevate
competenze in ambito FPE) come partner in grado di soddisfare
ogni esigenza tecnico/normativa applicando una metodologia innovativa basata non solo sulla profonda conoscenza di standard e
normative ma anche su capacità di Virtual Prototyping allo stato
dell’arte e, quando necessario, sulla
validazione sperimentale dei modelli e
delle scelte progettuali.
Le esperienze pregresse di EnginSoft
ed Idesa hanno messo in luce come le
problematiche legate alla lotta al fuoco non possano essere affrontate da
punti di vista singolari ma necessitino
un criterio di integrazione tra più discipline (system engineering, CAE, compliance, etc.). Con queste premesse e
con le competenze presenti in Engin@
Fire, le due società sono confidenti
che il nuovo brand diventerà un punto
di riferimento per la concretizzazione
di questa metodologia di approccio
nell’ambito della Fire Engineering.
Per ulteriori informazioni:
Marco Perillo, EnginSoft
[email protected]
Software update
INGLASS: oltre la camera calda
Sin dalla sua fondazione, la filosofia della INglass è stata quella di
investire nelle migliori tecnologie, ed in questa ottica abbiamo
scelto ANSYS poichè, dalle analisi fatte della concorrenza, lo
riteniamo essere la migliore soluzione per le nostre esigenze. La
facilità di utilizzo delle ultime versioni ANSYS Workbench ci ha
ulteriormente convinto di aver fatto la scelta giusta orientandoci
verso le soluzioni ANSYS. Inoltre EnginSoft ha dato prova di essere
un partner serio ed affidabile aiutandoci nella fase individuazione
del pacchetto di licenze a noi più utile. Nel complesso siamo soddisfatti della scelta da noi operata grazie alla quale aumententeremo
la qualità dei nostri prodotti.
Il gruppo INglass-HRSflow di San Polo di Piave viene fondato
da Maurizio Bazzo nel 1987 con il nome di Incos, per la progettazione e la costruzione di stampi per il settore delle materie
plastiche. Nel 1991 inizia a focalizzarsi e a specializzarsi negli Ing. Gianmatteo Bernardello
stampi rotativi multi-colore e multi-componenti per fanaleria Responsabile Ufficio Tecnico Stampi - Inglass
nel settore auto.
Nel 2001 nasce la divisione HRSFlow per la progettazione e la
realizzazione di sistemi a canale caldo per lo stampaggio ad iniezione
prodotto ai nuovi mercati emergenti ad alto potenziale di crescita,
di materiale plastico. A partire dal 2004 ingenti risorse sono state
dalla Cina, all’India, al Vietnam.
dedicate allo sviluppo della tecnologia di inietto compressione per
Nel 2010 viene diversificata ulteriormente la gamma di prodotto con
la produzione di ampie superfici in policarbonato in sostituzione del
l’introduzione della linea Multitech, dedicata allo stampaggio di comvetro nel settore automobilistico (Plastic Glazing).
ponenti con pesi ridotti e spessori fini che appartengono a settori
Nel 2007 l’azienda introduce un servizio in grado di ottimizzare il
quali medicale/packaging/chiusure/automotive o in generale per tuttempo di raffreddamento degli stampi tramite un’analisi di Cooling e
te le applicazioni che richiedono tempi ciclo veloci e rese estetiche
l’eventuale realizzazione di inserti speciali ottenuti con la tecnologia
molto elevate.
del Selective Laser Melting, nota anche come fusione laser di polveri
Nel 2012 INglass continua la sua crescita, aprendo nuove filiali tecmetalliche.
nico-commerciali, dall’India alla Colombia, alla Romania, per fornire
Nel 2009 si decide di accogliere una grande sfida sul mercato asiatiai propri clienti un servizio disponibile 24/7 in Europa, Asia, nelle
co e viene inaugurato un nuovo stabilimento produttivo ad Hangzhou,
Americhe, in Oceania e in Africa.
nei pressi di Shangai. Il nuovo stabilimento che ha una superficie
INglass si impegna a non essere un semplice fornitore di camere
totale 12.200 mq. totali, di cui 9.600 dedicati all’area produzione, ha
calde e stampi, ma un partner che segue il cliente durante tutto il
chiuso il 2012 con un fatturato pari a 12,5 Mil Euro. Lo stabilimento in
processo produttivo, fornendo soluzioni complete per la produzione
Cina produce e progetta sistemi a canale caldo replicando il modello,
di manufatti plastici. L’azienda, con il suo team altamente qualifila tecnologia e gli standard qualitativi della casa madre italiana. Il
cato, non solo fornisce servizi di ingegneria per la realizzazione di
mercato a cui si rivolge è quello asiatico, permettendo di fornire il
manufatti plastici, ma mette anche a disposizione il suo know-how
Testimonial
Newsletter EnginSoft Year 10 n°4 - 42
e le più avanzate innovazioni tecnologiche per seguire al meglio il
cliente, dalle simulazioni iniziali allo stampaggio del prodotto finito,
all’assistenza nel post vendita.
Gli sforzi sono diretti verso una costante ottimizzazione a livello
globale dell’organizzazione interna per lo snellimento dei processi.
A tale scopo l’implementazione dei sistemi di automazione ci ha
permesso di garantire riduzione drastica dei tempi di risposta, trasparenza e condivisione delle informazioni, a supporto del business
nostro e dei nostri partners.
INglass lavora per migliorare il vantaggio tecnologico attraverso ingenti risorse dedicate alla ricerca e sviluppo per garantire al cliente
ottime prestazioni in termini di qualità del pezzo stampato, risparmio
energetico, riduzione scarti.
L’utilizzo di ANSYS Workbench nella progettazione
L’utilizzo del software ANSYS WB prevede la verifica delle componenti dello stampo sotto l’effetto della pressione di iniezione e delle
forze di chiusura esercitate dalla pressa. Normalmente INglass valuta
la freccia della figura stampante che, se troppo elevata, può creare
spessori della lente fuori tolleranza e conseguenti difettosità. In base
al risultato dell’analisi viene valutato se e dove intervenire per supportare la figura e limitare il problema. Altra verifica svolta è quella
delle tensioni su alcune parti dello stampo ritenute critiche, come le
componenti dei movimenti, in cui la presenza di una cricca porta a
costosi e lunghi interventi di sistemazione e dunque di fermo della
produzione.
Seminario sull’ottimizzazione in
Università di Genova
Nell’ambito del Programma Accademico di EnginSoft per diffondere metodologie e tecnologie innovative presso Atenei e Centri di Ricerca Italiani abbiamo tenuto in Ottobre
un seminario sull’ottimizzazione all’Università di Genova, con la collaborazione del prof.
Viviani del Dipartimento di Ingegneria navale, elettrica, elettronica e delle telecomunicazioni (DITEN).
L’ing. Urban ha illustrato a professori e ricercatori i concetti di base e lo stato dell’arte sulle
metodologie di ottimizzazione messe a disposizione nell’ambiente modeFRONTIER evidenziando le potenzialità dello strumento nel combinare avanzate tecniche matematiche
alla ricerca applicata nel campo dell’ingegneria. Durante l’intervento sono stati presentati
alcuni casi applicativi industriali dimostrando come questi strumenti sono in crescente
diffusione in progettazione, ricerca e sviluppo a livello mondiale.
modeFRONTIER è una delle tecnologie di punta con cui aziende di livello internazionale ed
EnginSoft concorrono ad ottimizzare i propri prodotti e processi.
Per maggiori informazioni
Lorenzo Benetton - EnginSoft
[email protected]
43 - Newsletter EnginSoft Year 10 n°4
Testimonial
WIN-shoes: When Innovation makes Shoes
WIN-shoes intende mettere a punto un sistema integrato
ICT che permetta una totale rivoluzione nella organizzazione e gestione del lavoro nel mondo della industria manifatturiera toscana della calzatura.
In particolare, si intende mettere a punto una piattaforma
integrata che permetta di informatizzare il processo di progettazione e prototipazione del comparto, ad oggi quasi
totalmente artigianale, compiendo un efficace trasferimento tecnologico da settori dove sistemi CAS/CAD/CAE/CAM
vengono già da tempo utilizzati. Come originale elemento
di innovazione, WIN-Shoes intende inserire i parametri di
comfort come elementi determinanti la progettazione del
nuovo prodotto, prevedendo anche la messa a punto di un
sistema automatico di rilevazione oggettiva del comfort di
una calzatura tipo “calzino sensorizzato”.
Fig. 1 - Modello FEM di una calzatura “collaudato” da una modella virtuale
Ad oggi il processo di progettazione e realizzazione di
una nuova calzatura è molto artigianale e “manuale”:
WIN-Shoes si propone di innovare radicalmente questo
approccio, mettendo a punto un sistema ICT che permetta di simulare virtualmente la realizzazione di una scarpa la più “confortevole” possibile, così poi da andare in
prototipazione reale solo per assetti della scarpa già virtualmente ottimizzati, che saranno testati con innovativi
sistemi sensorizzati (essi permetteranno al software di auto-imparare, mantenendosi sempre “aggiornato” rispetto
alle soluzioni più promettenti e innovative individuate
dalle aziende) riducendo tempi e costi di realizzazione.
WIN-shoes propone inoltre un modello organizzativo
esportabile come elemento di innovazione del comparto:
la costituzione di una RETE d’imprese, gestite per capitalizzare il know how aziendale e per fare dell’innovazione il
motore a garanzia della competitività del comparto, grazie
Fig. 2 - Laboratorio di un calzaturificio
Research and Technology Transfer
Newsletter EnginSoft Year 10 n°4 - 44
alle figure strategiche del manager di rete e del manager di innovazione.
Un laboratorio dimostrativo WIN-shoes sarà il luogo privilegiato per la
formazione del personale sui nuovi strumenti ICT WIN-shoes e per la
divulgazione dei risultati di progetto.
Elementi di Innovazione del progetto
• Innovazione Organizzativa e gestionale di processo: il progetto
“forma” e “informatizza” il partenariato rivoluzionando il processo di progettazione e prototipazione dei prodotti, dal disegno
dello stilista ai file dei costituenti della calzatura.
• Innovazione Tecnologica di processo (progettazione e industrializzazione): creazione di una Piattaforma software di progettazione della calzatura: dai file ad una Calzatura Prototipale collaudata
da una Modella Virtuale in grado di esprimere il
COMFORT.
• Innovazione nella Struttura organizzativa e manageriale di comparto: le aziende si associano per fare
“struttura (viene introdotta la figura del manager
di Rete”) e si adotta la Filosofia della “innovazione
continua” (viene introdotta la figura del manager di
innovazione).
• Innovazione delle Strategie di comunicazione e
marketing: verrà creato un Laboratorio Dimostrativo
WIN-Shoes dove si concilieranno tradizione artigiana e High Tech. Qui si potranno toccare con mano
i prodotti della innovazione WIN-Shoes e saranno
organizzati Corsi di progettazione e utilizzo della
piattaforma WIN-Shoes
Prodotti attesi dal progetto
• Piattaforma informatica di progettazione e prototipazione della calzatura: tale piattaforma unitamente a device avanzati di prototipazione rapida permetterà operazioni di Virtual prototyping, Digital
manifacturing, Controllo di processo e prodotto,
nell’ottica di una riduzione del “Time to Market”
e dei Costi.
• Laboratorio Dimostrativo WIN-Shoes, dove troveranno spazio e utilizzo la piattaforma ICT WIN-Shoes, stampanti 3D, scanner 3D, ecc….
• Calzino sensorizzato, che permetterà la rilevazione
del comfort di una calzatura, in modo oggettivo e
riproducibile.
Partner di Progetto
Calzaturificio Everyn (capofila dell’iniziativa)
Calzaturificio Maruska (partner)
Tuscany Services srl (partner)
45 - Newsletter EnginSoft Year 10 n°4
Subcontractor
Laboratori ARCHA
Clinica Ortopedica dell’ Università di Pisa, U.O. Ortopedia e
Traumatologia II- Università di Pisa
Istituto di BioRobotica - Scuola Superiore Sant’Anna
EnginSoft Spa
Il progetto WIN-shoes è co-finanziato dalla Regione Toscana
POR CREO FESR 2007-2013
Per ulteriori informazioni:
Angelo Messina, EnginSoft
[email protected]
Fig. 3 - Operaio al lavoro in un calzaturificio
Fig. 3 - Architettura della piattaforma informatica
Research and Technology Transfer
MUSIC Project – First Review Meeting
“MUlti-layers Control&Cognitive System to drive metal and
plastic production line for Injected Components”
Eibar (Spain) September 18th-20th, 2013
The first review meeting of the MUSIC Project was hosted in the
new I4K – TEKNIKER building, in Eibar, where all partners gather to
report on their activities to the European Officer and to the appointed
European experts as project reviewers. Three “examiners”, related
to both private companies and public institutions, were required to
evaluate the project progress from a technical and scientific point
of view in order to verify the project assessment and the results
achieved along the first year of activity. This time frame is usually
the most critical one, since the project ideas, estimations and
plans have to be brought into real facts and actions, proving that
the project approach is solid and the final targets are realistically
achievable and very clear in mind.
The central topic of the MUSIC project, that of creating and using a
cognitive system able to analyze, control and predict the quality of
High Pressure Die Casted components as well as Plastic Injection
Molded parts, has required, along this first year, a considerable
research activity in order to provide a clearer view on needs and
expectations of the industrial sectors, investigating the most
common and significant defects and understand the corresponding
priorities, in relation to cost and energy saving, higher quality and
minimized scrap in HPD and PIM processes thanks to intelligent
and agile manufacturing.
Work-packages 1 and 2, respectively mainly focused on product
and process requirements and data acquisition&management have
been successfully reported to the experts. The results of a survey
concerning the HPDC sector, carried out in collaboration with
the European StaCast Project (Grant n. 319188, FP7-NP-2012CSA-6), have been presented, as well as the planning of a similar
activity to be managed with reference to PIM processes.
Starting from this primary research, the next step will be the
development and integration of a completely new ICT platform,
Research and Technology Transfer
based on innovative machine learning system linked to real time
monitoring, that will allow the active control of the quality of High
Pressure Die Casting (HPDC) of light alloys and Plastic Injection
Moulding (PIM) process.
The reviewers and officer clearly expressed the positive evaluation
of first period and their great expectations for the project future
evolution. We don’t forget our claim and we are proud to work so
that MUSIC becomes a Symphony in Smart Factories.
The MUSIC Projects invites you to support its initiative
concerning the survey related to the Plastic Injection
Moulding sector.
Visit the MUSIC Project website for further information
concerning the project and to fill in the questionnaire, that aims
at interviewing the EU industrial world to better understand
how project outcomes and expected impacts could provide
benefits and improve knowledge to be competitively applied
for all potential customers at different levels (SMEs, Industries,
Academia). http://music.eucoord.com/news/body.pe
All the collected information will be anonymously processed
in view of the targets of MUSIC Project. The results of statistics
elaboration of data will be made available to all the companies
answering the questionnaire.
For more information, please contact us at survey.
[email protected]
MUSIC at the International CAE Conference
The R&D area of the Conference hosted 14 running projects
representing the different application sectors in which EnginSoft
is operating, also thanks to European funding. It has represented
an important opportunity to show the Project to a wide international
and highly professional audience and some first prototypes were
displayed.
Newsletter EnginSoft Year 10 n°4 - 46
CLINIC OPTIMIZER
Interactive visualization for your
personal and intelligent choice
of medical treatment
Recently, the RWJF Hospital Price Transparency Challenge
was released. The challenge aims at increasing the
understanding and use of recently released hospital
price data. The “visualization” category of the challenge
encouraged submissions that allow users to better
understand aspects of the data. ClinicOptimizer is a
submission by Lionsolver Inc. It consists of an interactive
visualization that lets a customer select the best clinic
based on his or her condition and preferences.
Fig 1. The ClinicOptimizer software powered by Lionsolver Inc.
Picking a favorite hospital to be treated in after receiving a
diagnosis is usually a difficult task. This is a very concrete
example of multiple-objective optimization.
Ideally, one would like to maximize at least the following
variables:
1. Cost of treatment
2. Quality of treatment
3. Proximity to home
Fig 2. The three charts above show how the cost (top), quality (middle) and size (bottom) of the clinics
available for your treatment are distribuited across the nation
In practice, one is dealing with trade-offs.
Whilst ClinicOptimizer will not solve all the issues involved
in this critical decision, it will provide a more quantitative
input to the choice. The results are based on publiclyavailable information: improvements in the availability of
such data (provided that privacy issues are appropriately
addressed) will increasingly empower individual patients in
their dealings with profit-seeking entities such as insurance
companies and hospitals.
For more information:
Roberto Battiti, Reactive Search
[email protected]
47 - Newsletter EnginSoft Year 10 n°4
Fig 3. Ranking of preferred clinics according to the cost/quality tradeoff choosen by the user
In Depth Studies
International CAE Conference 2013
Global focus on the future
Simulation: the new design challenges
When projects turn into numbers and data become sequences able
to predict any single variable; when complicated calculations are
able to work out any step of any process thanks to increasinglycapable computers and software, then we understand the power of
computer-aided simulation. Even if we don’t realize it, simulation is
deeply rooted in our everyday life. The creation of computational
models allows us to obtain more precise and reliable results with a
dramatic reduction of time, economic resources and risks. Virtual
prototyping has already become an essential reality – no longer
confined to the laboratory, but a necessary component of every
industrial sector.
CAE Conference: a world-wide focus on the future
This renowned engineering event in the heart of Europe brought
together scientists, researchers and managers from various
sectors to show how “engineering simulation” can improve our
tomorrow. They shared their progress in medicine, engineering,
environmental-friendly and renewable energy, safety, aerospace
exploration and the management of natural phenomena; together
exploring how we may improve our future, our world and our lives.
That the answer lies (at least in part) through the blossoming of
technological advancement was made visible and almost tangible
over the two days of the International CAE Conference in Pacengo
sul Garda from the 21st to the 22nd October. Our prestigious
conference attracted a global audience from various backgrounds
and disciplines with a common understanding that no product,
process or service innovation can disregard engineering simulation.
This year (the 29th) extends an incredible record. The Scientific
Director of the International CAE Conference, Stefano Odorizzi,
initiated this remarkable event thirty years ago, and has nurtured
and promoted it over three decades with the aim of capturing the
state of the art of a methodology whose great potential has proved
to be absolutely revolutionary.
CAE Conference
Fig 1 - An exciting moment of the Conference: the connection with the Italian astronaut
Parmitano directly from the Space Station during the speech of Maurizio Cheli
What started as an Italian initiative is now not just a European, but
also a worldwide reference point. Delegate numbers confirm this
trend: a 25% increase in participants (in comparison to 2012),
with thirty seminars and workshops divided into different sectors
(energy, civil engineering, transportation and biomechanics,
among the others), this year with a strong focus on Aerospace &
Defense. Fourteen projects displayed their results and prototypes in
the R&D area, and 30 remarkable sponsors (including Nvidia, HP,
Newsletter EnginSoft Year 10 n°4 - 48
IBM and ANSYS) filled our exhibition
space - including EnginSoft, the Italian
multinational company, headquartered
in Trento, founded by Stefano Odorizzi
in 1984.
Another Italian representative, contributing
to another space mission, is Paolo Belluta
who works at the NASA Jet Propulsion
Laboratory in Los Angeles. He flew back
to Italy to participate in the International
CAE Conference and presented his
work as driver of the Mars rovers Spirit,
Opportunity and now Curiosity. “These
vehicles cost millions of dollars and
could be lost forever due to a simple
accident. If we couldn’t take advantage
of engineering simulation to analyze the
driving context and condition, we would
be the ones totally lost!”
EnginSoft is renowned for its computer
aided engineering systems and
solutions, its commitment to the value
of virtual prototyping and the priority
it places on innovation, dissemination
and education in the latest innovative
methods and software. For these reasons
EnginSoft has consistently prioritized
the development of the International
CAE Conference (www.caeconference.
com), as well as its quarterly EnginSoft
Non-seismic systems and green
SBES (Simulation Based Engineering
nuclear energy
& Sciences) Newsletter. These tools
The International CAE Conference was
honored by the presentations of two
offer simulation engineering experts
and opinion leaders the ability to share
renowned Italian scientists, Francesco
Iorio and Carlo Sborchia, authors of
information, projects and experiences
Fig
2
Prof.
Odorizzi
from
EnginSoft
during
the
Conference
welcome
extraordinary projects, well-supported
throughout the year, as well as enjoying
the opportunity to gather in person at the
by simulation. Francesco Iorio, engineer
annual Conference.
and professor at Politecnico di Milano, is
the creator of the futuristic non-seismic damping system of the
This year we were joined by extraordinary key-note speakers
Isozaki Tower in Milan. In 2015 the tower will reach a height of 207
Maurizio Cheli, the Italian astronaut; Alexander Simpson, Global
meters to become the tallest skyscraper in Europe. Carlo Sborchia
Research leader of General Electric; Catherine Riviere, President
is an Italian genius working in France in the ITER project, the
of PRACE and GENCI general manager; Michael Gasik, from Aalto
revolutionary nuclear reactor that, using hydrogen (tritium) instead
University Foundation in Finland and Bernardo Schrefler, who is
of uranium, will solve the problem of radioactive waste.
involved in Padova and Houston in medical research, to predict
cancer evolution with relation to medical treatments.
Investing in young resources and technologies
That’s how we could get out of the crisis
Aerospace in the forefront: from Shuttle mission to Mars
The economic difficulties that characterize the industrial world are
Maurizio Cheli, the Italian astronaut with over 360 hours of space
strongly affecting scientific research. Laboratories and universities
flight, including the Space Shuttle, and 4500 hours flight on highhave their funds reduced, with the result that innovative projects
performing aircraft, opened the International CAE Conference
2013 and sealed a connection between this event and the Torino
cannot achieve their potential impact – this disproportionately
Piemonte Aerospace initiative that took place at Lingotto from
affects younger researchers who have form the core teams of
such projects. For this reason the International CAE Conference
October 22nd to 24th.
has introduced the Poster Award aimed at Research Centers and
This is the first step of a new collaboration between the companies
working in the aerospace sector and the scientific and institutional
Academia with an impressive feedback of over 300 projects related
to simulation engineering. Forty of them have been selected for
community. “The technology coming from aerospace and
judging by the scientific committee, with five prizes awarded. The
applied to everyday life can strongly support the Italian economy.
Access to supercomputers is essential to guarantee industrial
most innovative project has also been rewarded with an unexpected
opportunity for its young author, the proposal is to be employed
competitiveness and top-level scientific research. From this
by EnginSoft: “It’s a personal commitment that I have decided to
perspective, Europe has no reason to be envious of United States
and Asia,” stated Cheli.
make,” observes Stefano Odorizzi, “reflecting an observation that
in Italy there’s little place for meritocracy and a serious lack of
It is rare for a conference presentation to generate a widespread
opportunities for talented people. Companies don’t pay attention
emotional impact on an audience, but this was certainly the effect
of a live connection with International Space Station and the ESA
to them, while managers should be more responsible and act
astronaut Luca Parmitano. He greeted the audience by reminding
accordingly. I hope to become a positive example to be followed,
us how essential simulation is to the support of space mission
and if other Italian companies would do the same, investing in
preparations, allowing astronauts to train and cope with extreme
young talent, the way-out of this crisis would be much closer.”
and risky situations. “I have to thank all you researchers”, he
For more information: www.caeconference.com
admitted, “for enabling me to be here and talk to you!”
49 - Newsletter EnginSoft Year 10 n°4
CAE Conference
CAE Conference 2013: will ANSYS Mechanical
Release 15 satisfy technical user expectations?
Discussion and final considerations at the ANSYS
Mechanical meeting
Every year, at the conclusion of the International CAE Conference,
the use of ANSYS Mechanical by Italian engineers is reviewed.
Many interesting inferences may be made from considering
parameters such as the number of attendees, quality of technical
presentations and, from the software provider’s viewpoint, the
ability of the latest release to satisfy the technical expectations of
each participant.
This year there were 103 attendees during the ANSYS Mechanical
session, the vast majority remaining until the final presentation.
The overall conclusion was of a positive correlation between users’
demands, code capabilities and the technical requirements of the
engineering challenges presented.
The “leitmotif” of these presentations is their organization around
individual special features of the ANSYS code: each presenter is a
veteran ANYS user who has been chosen because of their ability
to present a detailed technical solution and the benefit that was
obtained.
The first speech focused on the general state of art of the ANSYS
Mechanical code and the development guidelines that were
introduced by ANSYS to ensure the spirit of renewal which
characterizes the continuous improvement of the code in spite
its ‘seasoned’ technology. Five pillars form the basis of these
developments, leading to release 14.5 and further exposed in
released 15. These are HPC (High Performance Computing)
technology, ACT (the ANSYS Customization Tool), significant
improvements in meshing performance, multi-physics, and new
technologies in fatigue and fracture mechanics.
Ing Santucci from Ansaldo and ing Perna from EnginSoft showed
their presentation on huge use of HPC, revealing an innovative
software-hardware structured approach that has already made a
major impact on Ansaldo productivity.
CAE Conference
Ing Cova from Sacmi demonstrated the utilization of ACT to
incorporate a customized tool into the ANSYS environment to deal
with specific issues in ceramic material behavior.
Ing Mechi of Continental has introduced an interesting discussion
about ANSYS capabilities in fatigue analysis, with reference to the
features of the nCode software, raising various questions about its
detailed usage.
Ing Biondi of Ansaldo presented an interesting analysis of a turbo
alternator that linked a determination of its magnetic field with the
resulting structural stress field analysis arising from the resulting
Maxwell forces.
Other presentations from ing. Gardi from Cira, ing Bistolfi from
Franco Tosi Meccanica, ing Raffaelli from INFN and ing Pacieri
from Umbra Cuscinetti shared some common points characterized
by challenging meshing requirements and demanding algorithm
expectations. Together, they demonstrated that the powerful
capabilities of ANSYS in each of these areas have enabled them to
identify good technical solutions during the development of their
products.
Finally, as the ANSYS Technical Organizer, I want to thank all the
attendees and speakers for their reciprocal contributions. I would
further like to remind you that we will continue to share with you
all of the continuing and dynamic innovation within the ANSYS
product family through our extensive programme of webinars and
courses in our different competence centers. Please consult the
plans and events publicized on our website www.enginsoft.it/eventi
For more information:
Roberto Gonella – EnginSoft
[email protected]
Newsletter EnginSoft Year 10 n°4 - 50
Scilab at the International CAE Conference 2013:
what a great session!
The Scilab Session and Workshop at the 29th edition of the International
CAE Conference was a great success!
Scilab is the worldwide Open Source reference in numerical computation
and simulation software and it has been adopted in all the major strategic
and scientific areas of industry and services, such as aerospace,
automotive, electronics, energy, defense and finance.
The session hosted speakers from all over Europe and India, and the
presentations covered both academic and industrial applications.
Lea Florentin and Eloy Crespo, from Eramet Research, proposed two
approaches to model metallurgical reactors with Scilab. Metallurgical
reactors are characterized by their complexity, especially in terms of
chemistry, heat transfer, transport phenomena, and thermodynamics;
they showed how modeling such systems can lead to an improved
understanding of the process and offer the opportunity to optimize the
performance of the reactor.
There were two excellent presentations from the field of civil engineering:
Sanjeev Gahlot, Govt. India, illustrated an interactive Scilab program
for the simulation and the optimization of a 2-dimensional Truss, while
Sukumar Baishya, Associate Professor at the North Eastern Regional
Institute of Science and Technology, presented a program for the analysis
of the allowable bearing pressure of shallow foundations in layered
soil. He explained that, on many occasions, the detailed geotechnical
characterization of a site may not be possible due to difficult ground
conditions and limited time and resources. However, the developed Scilab
program is able to utilize input data gathered during site explorations to
compute the Bearing Capacity and the Allowable Bearing Pressure and
generates an appropriate geotechnical report.
The first speaker from the academic world was Gabriele Santin, from the
Padua University, who explained that the theory of Radial Basis Functions
(RBF) is of growing importance in the field of approximation, especially
when dealing with data coming from scattered samplings in highdimensional spaces. Nevertheless, in certain conditions this method can
be unstable and can suffer from ill-conditioning: hence he presented a
Scilab implementation of some tools for the fast and stable computation
of RBF approximants in a wide class of problems. Giovanni Conforti, from
the Berlin Mathematical School, introduced the main tools that Scilab
51 - Newsletter EnginSoft Year 10 n°4
offers to tackle problems arising from stochastic modeling, describing
and implementing a stochastic numerical algorithm to solve elliptic PDEs
with a special focus on the heat equation. His algorithm is based on the
celebrated Feynman-Kac representation formula. The last talk of the
morning session was presented by Davide Poggiali, Padua University, and
it concerned the resolution of a gamma camera in a clinical setting. From
a diagnostic point of view, it is useful to know the expected resolution of
a gamma camera at a given distance from the collimator surface for a
particular setting in order to decide whether it is worth scanning patients
with small lesions and to make appropriate corrections. He created
a package to obtain the theoretical resolution of a gamma camera at
different distances and compared these values with experimental results.
In the afternoon, the Openeering team and Jocelyn Lanusse, from Scilab
Enterprises, led the Workshop. The main themes concerned the latest
Scilab version, how Scilab can be used for building comprehensive
industrial applications using external modules and the Return On
Investment using Scilab. The workshop lasted for two hours and the
attendees took an active part in the discussion, carefully dissecting
each topic. The Workshop was also the setting for the official launch of
the Scilab Black Belt Course, which the Openeering team has carefully
designed to cover the full spectrum of Scilab features whilst keeping it
compact and yet enriching it with meaningful examples.
Moreover, this session also had the pleasure of hosting an introduction
to the “Mathematical Desk for the Italian Industry,” whose mission
is to build a bridge of common interest between the Italian scientific
community of applied mathematics and the world of Italian enterprise.
The cooperation envisaged is focused in particular on the development
of industrial research projects, possibly in the international context of
European networks, through effective mediation in the field of scientific
and technology transfer based on the role of mathematics.
The Scilab session will return to the International CAE Conference in
2014, where we look forward to meeting you again: stay tuned!
For more information:
Anna Bassi – Enginsoft
[email protected]
CAE Conference
CAE Conference 2013 sessione MAGMA:
un grande successo
La CAE Conference 2013 ha chiuso i battenti confermando, anzi aumentando il già notevole successo degli anni precedenti.
All’interno dell’evento fra le numerose sessioni parallele, che hanno
permesso approfondimenti nei vari settori di applicazione virtuale,
è stato possibile assistere anche alla sessione specifica MAGMA,
software dedicato all’analisi virtuale dei processi produttivi di fonderia. In sintonia con il risultato complessivo dell’evento, anche la sessione MAGMA è stata caratterizzata da una importante affluenza di
ascoltatori non solo appartenenti al settore specifico delle fonderie.
In particolare quest’anno è stato dato ampio spazio agli utilizzatori
che hanno esposto la loro esperienza nell’utilizzo dei sistemi virtuali
nella logica di integrazione Processo-Prodotto.
cluso la sessione illustrando l’evoluzione del software che dal prossimo anno vedrà il rilascio della versione definitiva dell’Ottimizzatore
che verrà integrato nel modulo base nella release 5.3, insieme a molte altre novità sicuramente molto utili, come ad esempio il controllo
automatico del livello del battente di colata nel bacino, per i processi
di colata in gravità.
Ancora una volta dobbiamo ringraziare tutti i relatori che hanno contribuito attivamente al successo di questa edizione e tutti i partecipanti che con il loro intervento hanno reso vivace la nostra sessione.
Un ringraziamento particolare lo dobbiamo al Dr. Sturm e al Dr. Bramann che hanno mantenuto un contatto diretto con gli utenti Italiani,
Gli argomenti trattati sono stati molteplici comprendendo non solo le attività specifiche di casting negli ambiti
di produzione dei materiali Ferrosi e non Ferrosi, ma andando anche a considerare l’attività produttiva delle anime in sabbia, con preciso riferimento all’ultimo modulo
nato in casa MAGMA: C+M (ossia Core and Molds).
Proprio relativamente al modulo C+M è stato possibile assistere a 2 presentazioni condotte rispettivamente
dall’Ing. Timpano della Agusta di Benevento, il quale ha
esposto il lavoro di progettazione e di realizzazione di
un’anima piuttosto complessa per la produzione di una
scatola ingranaggi per un motore per elicotteri, e dal Sig.
Stefano Tamelli della Modelleria Brambilla di Correggio,
il quale ha esposto un caso di produzione di un’anima del
circuito di raffreddamento ad acqua di una testa cilindri
per un motore benzina di nuova generazione.
Tra le aziende che hanno partecipato attivamente alla sessione ricordiamo: Modelleria Brambilla, Acciaierie Fonderie Cividale, iGuzzini,
FAS Fonderia Acciai Speciali, Agusta, Fonderie Mario Mazzucconi,
StaCast Project, Università di Padova, ABOR.
Al termine delle memorie presentate dagli utenti è stata la volta del
Dr. Sturm, Direttore Generale della MAGMA GmbH, il quale ha con-
CAE Conference
confermando come MAGMA GmbH sia al fianco delle fonderie per
rendere sempre più efficace l’uso della simulazione in questo importantissimo settore manifatturiero.
Per ulteriori informazioni:
Giampietro Scarpa, EnginSoft
[email protected]
Newsletter EnginSoft Year 10 n°4 - 52
Forge NxT: l’Italian User Meeting raddoppia
L’annuale appuntamento della CAE Conference quest’anno ha visto una
partecipazione nazionale ed internazionale ancora maggiore; l’importante contributo delle numerose sessioni specifiche ha permesso approfondimenti interdisciplinari di alto valore. Tra esse si è distinta la duplice
sessione dedicata a Forge & ColdForm: il 21 Ott. è stata presentata in
anteprima la prossima release del SW Forge, il 22 Ott. si sono avvicendati gli interventi di utilizzatori e della casa madre: Transvalor.
Le novità presentate sono lo specchio della roadmap di sviluppo del SW
Forge, ed hanno dato i punti chiave ed i vantaggi di utilizzare le nuova interfaccia e le new features inserite: new mesh, Induction Heating,
nuove cinematiche, etc. Alla presentazione dell’Ing. Andrietti - Software Production Manager - è seguita una sessione hands-on con ottimo
coinvolgimento dei numerosi intervenuti, i quali hanno potuto testare la
bontà delle innovazioni apportate, contribuendo con interessanti spunti
al continuo sviluppo del SW stesso e della sua interfaccia grafica, che lo
rende ancora più intuitivo, robusto ed user-friendly.
Il nuovo modello di calcolo per l’Induction Heat Treatment permette di
analizzare non solo la distribuzione di temperatura durante il processo, ma anche l’evoluzione metallurgica, aumentando la precisione dei
risultati. Grande attenzione è stata dedicata dai più di 50 partecipanti
alla Best-Practice Heat-Treatments & New features in steel quenching
simulation, della quale i punti salienti sono la possibilità di modellare il
processo di cementazione, austenitizzazione, tempra e la nuova possibilità di importazione, esportazione e visualizzazione dei dati.
Il coinvolgimento degli utilizzatori ha sottolineato l’alto valore aggiunto
che l’uso del FEM e del CAE apporta ai processi di manufacturing, come
evidenziato nello studio Analysis and optimization of heating process
for large forgings quenching through finite elements analysis, presentato dal’Ing Curbis - PhD UNIPG presso Società delle Fucine di Terni.
L’ottimizzazione del processo di riscaldo di grandi fucinati (cilindri di
laminazione >250ton) destinati alla tempra differenziale permette di
massimizzare la qualità del prodotto forgiato, incrementando la durezza
dello strato di lavoro, il cui spessore varia in funzione delle specifiche
richieste. La simulazione numerica favorisce un approccio sistematico e
scientifico alla risoluzione di tali problematiche industriali, specialmente
quando il rapporto di scala sperimentale-reale è molto elevato.
In ambito di materiali innovativi è stato molto apprezzato lo studio: Dif-
53 - Newsletter EnginSoft Year 10 n°4
ferences in Forging Process for a Component Type “Convergent Venturi”
Evaluated in Different Material, condotto dall’ing. Caracciolo - Product &
Process Engineering Director - e dall’ing. Baruffaldi in Officine A. Melesi.
L’approccio FEM ha permesso di valutare in maniera preventiva la criticità del processo di forgiatura di un componente per OIL & GAS in INC
800H. La coerenza con i dati sperimentali e la robustezza delle analisi
virtuali hanno permesso di ottimizzare il processo minimizzando il materiale usato (-20%), con conseguente riduzione di energia e costi totali
pari al 30%. Altrettanta attenzione è stata rivolta allo studio di tesi in
ambito di deformazione di metalli non ferrosi: A Study of the Parameters’
Influence on Dies Resistance in Hot Forging trough Numeric Simulation,
condotto in Fonderia Maspero da Canesi - POLIMI - con la supervisione
dell’Ing. Di Modica. L’analisi ha identificato e quantificato i fattori di causa di rottura stampi nella produzione di un grande elemento strutturale
Automotive in alluminio (EN - AW 6082). Il risultato di minimizzazione
dello stress termo-meccanico costituisce la base per lo sviluppo di una
politica adeguata per la progettazione e la gestione delle attrezzature.
In ambito Cold Forging, l’ing. Zoppelletto - A.D. Zoppelletto SpA - e l’ing
Bassan - PhD UNIPD - hanno presentato lo studio di un processo di
stampaggio multistazione con transfer automatico di un particolare di
forma complessa e strette tolleranze dimensionali: Process analysis and
investigation of defects in multistage cold forging by using finite element
method. I risultati delle analisi FE si sono dimostrati affidabili ed in ottimo accordo con i dati sperimentali: previsione di difetti, bilanciamento
di forze ed ottimizzazione di processo, massimizzando la produzione e
minimizzando i costi.
Il centro di competenza deformazione di metalli di EnginSoft, approfondendo tutte le tematiche presentate, ha coordinato le numerose domande degli intervenuti e, attraverso un confronto di tutti i partecipanti, ha
collezionato tutti gli spunti di sviluppo.
Ringraziando gli intervenuti e tutti coloro che hanno contribuito, si rinnova l’appuntamento alla edizione 2014 dell’Italian Forge Users’ Meeting:
Ad Majora!
Per ulteriori informazioni:
Andrea Pallara, EnginSoft
[email protected]
CAE Conference
The International CAE Conference 2013 welcomed
participants from Japan
We had the pleasure to welcome and interview Mr. Motoaki Ioroi
from Japan who works for Honda Engineering Europe Ltd. in the UK.
from around the world can participate more easily. This would provide
even more opportunities for lively discussions.
A.Kondoh: Could you please tell us a bit about your work and
what type of CAE you are using at Honda Engineering?
A.Kondoh: Please tell me about your future vision for the use of
CAE in your company?
Mr. Ioroi: Since I joined the UK office a few months ago, I have been
gathering information to investigate the leading technology in Europe
for automotive powertrain components.
I had been using ADSTEFAN, a casting simulation system, for fluid
flow analysis and solidification analysis of hot metal in dies at our site
in Japan before I moved to the UK office. At the moment, my main
task is to study new technology, I don’t use CAE currently in the UK
office.
Mr. Ioroi: We anticipate that cast components will become more
complicated and therefore, the casting performance is going to be
more challenging in the future. At the same time, we are required to
reduce product development lead-time in order to cut costs. I hope to
establish the ideal casting process for the components with complex
geometries by applying CAE – This approach will indicate, will tell us
which dies can deliver products with the targeted quality and without
any errors.
A.Kondoh: What were the main objectives for your participation
at the CAE Conference?
This interview was conducted by Akiko Kondoh,
Consultant for EnginSoft in Japan
Mr. Ioroi: Our goals were to gather information about the CAE
development trends in Europe and the recent case studies of casting
CAE and forging CAE and to build networks with companies and
universities which are developing CAE software products. We think
that it is important to value the relationships between European
companies and universities so that we can take full advantage of our
CAE usage and improve the efficiency of product development.
A.Kondoh: What are your main impressions of the
CAE Conference?
Mr. Ioroi: The conference provided an atmosphere of great openness,
it was easy to discuss with others. The only pity was that I could not
understand the entire content of some Italian sessions. I hope that
next year there will be more English sessions so that the participants
CAE Conference
Left: Tomoya Tanaka of Honda Engineering Co.,Ltd.
Middle: Motoaki Ioroi of Honda Engineering Europe Ltd.
Right: Stefano Odorizzi, CEO of EnginSoft
Newsletter EnginSoft Year 10 n°4 - 54
CAE Poster Award 2013
An acknowledgment to young researchers’ creativity
The second edition of the award for “the International Poster Award:
A poster for CAE” has been really successful, both in terms of
participants and quality of the submitted works.
This initiative has been promoted and sponsored by EnginSoft,
being one of the promotion and dissemination activities that the
company is constantly committed in to foster simulation culture.
This award has a double aim: the first is to acknowledge the quality
and innovation of the project developed in the universities and the
second is to offer a privileged context in which academic experiences
and industrial world could meet and get mutually known.
41 posters, submitted by Italian and foreign universities and
research centers, passed the selection; 5 projects won the award
and 4 deserved the “mention of distinction”.
Posters were evaluated and voted by registered users and by
the Scientific Committee members, consisting of professionals
committed in transferring and disseminating numerical simulation
techniques and knowledge, both on academic and industrial level,
therefore able to influence the future of R&D: Aronne Armanini
(Università di Trento, Italy), Sanzio Bassini (CINECA, Italy), Roberto
Battiti (Università di Trento and co-founder of Reactive Search,
Italy), Franco Bonollo (Università di Padova, Italy), Gabriele
Dubini (Politecnico di Milano, Italy), Natalie Fedorova (ITAM SB
RAS, Russia), Giorgio Fotia (CRS4, Italy), Michael Gasik (Aalto
University, Finland), Carlo Gomarasca (Ansys, Italy) Gianluca
Iaccarino (Stanford University, USA), Giuseppina Maria Rosa
Montante (Università di Bologna, Italy), Enrico Nobile (Universita
di Trieste, Italy), Bernardo Schrefler (Università di Padova, Italy),
Christos Theodosiu (DTECH Corp., Greece) and Giorgio Zavarise
(Università del Salento, Italy).
The winners were celebrated on October 21st in the frame of the
International CAE Conference, held in Lazise (Verona). During the
55 - Newsletter EnginSoft Year 10 n°4
ceremony, presented by Luca Viscardi of Radio Number One, the
five best posters were officially announced in front of the audience
and their authors personally awarded with a tablet pc by Stefano
Odorizzi, enterprising and prominent expert of Computer-Aided
Engineering and Maurizio Cheli, astronaut, pilot, test driver and
successful manager.
The list of the five award winners is provided here below:
1. Design by Optimization of a Controllable Pitch Marine
Propeller
Stefano Gaggero, Michele Viviani - University of Genoa
2. Thermo-fluid dynamics model of two-phase system alloy-air
inside the shot sleeve in HPDC process
Roberto Meneghello - University of Padova
3. FEM Analysis, Modelling and Control of a Hexacopter
Angela Ricciardello, Valeria Artale, Andrea Alaimo, Cristina
Milazzo, Luca Trefiletti - University of Enna KORE
4. CFD characterization and thrombogenicity analysis of a
prototypal polymeric aortic valve
Filippo Piatti, Matteo Selmi, Alessandra Pelosi, Alberto
Redaelli - Politecnico di Milano - Thomas E. Claiborne, Danny
Bluestein - Stony Brook University, New York
5. Stenting in coronary bifurcations: image-based structural and
hemodynamic simulations of real clinical cases
Sebastian G. Colleoni, Stefano Morlacchi, Claudio Chiastra,
Gabriele Dubini, Francesco Migliavacca - Politecnico di
Milano
An in-depth perspective on each project is provided in the next
pages.
For further information, please go to www.caeconference.com,
Poster Award section, or contact [email protected]
CAE Poster Award
CAE POSTER AWARD 2013: WINNER
Stenting in Coronary Bifurcations: Image-Based
Structural and Hemodynamic Simulations
of Real Clinical Cases
The majority of current numerical models simulates typical stenting
procedures in idealized geometries. Consequently, such models can
only provide standard guidelines without specific indications for the
optimal interventional planning of each patient. The aim of this work
is the implementation of patient-specific structural and fluid dynamic
models that use image-based reconstructions of atherosclerotic
bifurcations. Particular attention is paid to the plaque identification
and the insertion of stents by simulating their advancement in the
artery. Two clinical cases involving a coronary bifurcations of the left
anterior descending artery have been investigated.
Materials and Methods
Image-based 3D atherosclerotic coronary bifurcation
The pre-stenting internal wall surfaces are generated from a combination
of conventional coronary angiography and computed tomography
angiography (Fig. 1, left). These surfaces have been used to construct
3D solid models of the two coronary bifurcations investigated (Fig. 1,
right). External wall surfaces were created choosing the diameters in
order to respect physiological values of the internal diameter and wall
thickness of the arterial branches investigated (LAD). The geometry
is discretized using a fully hexahedral mesh (Fig. 2a). Finally,
atherosclerotic plaques were identified based on the distance between
each node and the centerline of the external wall surface (Fig. 2b).
Stent models and prelinimary structural analyses
The two clinical cases simulated involve two coronary stents: the
Endeavor Resolute by Medtronic and the Multilink Vision by Abbott
Vascular. Their 3D CAD models and discretizations are shown in Fig.
3. To correctly position the stents in the complex arterial geometries,
crimping and advancement (Fig. 4) of the devices are simulated using
an internal guide following the vessel centerline.
Results
Simulated procedures
Numerical simulations of stent deployment were performed using
a commercial code as quasi-static processes. Two simulations
are carried out following the clinical indications provided by the
physician who performed the treatments at the Hospital Doctor Peset
in Valencia (Fig. 5 and 6). Initial stressed configurations of the devices
are imported from the preliminary analyses to guarantee a correct
positioning and accurate mechanical results.
Fluid dynamic simulations
The final geometrical configurations obtained are then used to perform
steady fluid dynamic analyses. Preliminary results (Fig. 10) highlight
the criticism of the overlapping region.
CAE Poster Award
Figure 1. Creation of the 3D geometries (right) of the two left anterior descending
coronary arteries investigated. On the left, image-based reconstructions of the prestenting internal wall of the vessel created using a combination of conventional
angiography and computed tomography
Figure 2. A) Hexahedral mesh of the geometry of case
2. B) Plaque identification is based on the comparison
between a typical radius of a healthy LAD and the
distance of each node to the centerline. If the distance
is lower than the radius (black arrows), the node will
be part of the plaque; otherwise (white arrow), it will be
part of the arterial wall
Figure 3. 3D models resembling
the two stents used: Endeavor
resolute (left) and Multilink
Vision (right). Dimensions and
discretization of their sections are
shown, too
Figure 4. Structural simulation of the stent advancement along a cylindrical guide
constructed following the post-angioplasty vessel centerline. Final stresses and
geometrical configurations have been used as a starting point for the final simulations
Biomechanical analysis: overlapping stents and
straightening of the artery
Main biomechanical results are shown in Figs. 7, 8 and 9 in terms of
stresses in the arterial walls, stresses in the stents and final geometrical
configurations. Overlapping of stents is proved to be a critical area due
to higher stresses and metal-to-artery ratios. Moreover, in both cases,
a straightening of the artery is found at the end of the procedure, in
accordance to in vivo measurements found in literature.
Newsletter EnginSoft Year 10 n°4 - 56
Stefano Morlacchi, Sebastian G. Colleoni, Claudio Chiastra, Gabriele Dubini,
Francesco Migliavacca - Politecnico di Milano
Ruben Cardenes, Ignacio Larrabide, Alejandro F. Frangi - Universitat Pompeu
Fabra and CIBER-BBN Barcelona
Jose Luis Diez - Hospital Valencia
Figure 6. - Steps of the technique performed for Case 2: A) pre-dilatation performed
with a 2.0 mm balloon expanded at 12 atm; B) a 28 mm long Multilink Vision stent with
nominal diameter of 3 mm is deployed at 14 atm across the bifurcation between the
LAD and its first diagonal branch; C) the procedure is ended with a post-dilatation at 18
atm in the proximal part of the main branch using a 3 mm balloon; D) final configuration
after recoil
Figure 9. Red and light blue shapes correspond to the pre-stenting surface and the poststenting geometrical configuration obtained with numerical simulation. Straightening of
the arterial wall is found in both cases. This occurrence is in accordance to the cited
publication where stented arteries are reconstructed using a combination of angiography
and IVUS
Figure 7. Maximum principal stresses contour maps of several sections along the main
branch of the coronary tree investigated in Case 1. Results are taken at the end of the
whole procedure. Absence of plaque and minimal expansion of the artery results in very
low stresses in the proximal area (left) while higher stress values can be found in the
distal part of the main branch and, particularly, in the overlapping area
Figure 7. Maximum principal stresses contour maps of several sections along the main
branch of the coronary tree investigated in Case 1. Results are taken at the end of the
whole procedure. Absence of plaque and minimal expansion of the artery results in very
low stresses in the proximal area (left) while higher stress values can be found in the
distal part of the main branch and, particularly, in the overlapping area
57 - Newsletter EnginSoft Year 10 n°4
Figure 10. Velocity field (top) and wall shear stress magnitude (bottom) contour maps
for a steady state simulation. Preliminary results prove that the overlapping area and the
walls opposite to the bifurcations are affected by wall shear stresses lower than 0.5 Pa
that is a critical value
CAE Poster Award
CAE POSTER AWARD 2013: WINNER
Figure 5. - Steps of the procedure performed in Case 1: A) pre-dilatation with a semicompliant 15 mm long angioplasty balloon with a diameter of 2.5 mm; B) deployment of
the 15 mm long Endeavor stent across the distal bifurcation inflating a 2.75 mm balloon
at 12 atm; C) positioning and dilatation at 14 atm of the second 15 mm long Endeavor
stent across the proximal bifurcation with a 3.0 mm balloon; D) final configuration after
recoil
Conclusions
This work shows the feasibility of implementing a patient-specific
virtual model replicating actual clinical cases. Standard medical
images (CCA and CTA) were used to create the 3D pre-stenting
geometry and the intervention was simulated following the clinical
indications provided. Moreover, simulations of crimping and insertion
were necessary to find the correct positioning of the devices in
complex image-based geometries. From a biomechanical point
of view, overlapping of stents has been recognized as a critical
occurrence due to modified hemodynamic and structural variables
both in the artery and the devices.
Acknowledgments: This work has been partially supported by the Italian
Institute of Technology (IIT, Genoa, Italy).
CAE POSTER AWARD 2013: WINNER
Design by Optimization of a Controllable
Pitch Marine Propeller
The propeller design is an activity which nowadays presents ever
increasing challenges to the designer, involving not only the usual
mechanical characteristics fulfillment (with maximum efficiency) and
cavitation erosion avoidance, but also other cavitation side effects,
such as radiated noise and/or pressure pulses. This is evident with
the ever increasing demand for improvement of comfort onboard and
concerns about radiated noise problems, especially in proximity of
protected areas or for Navy ships. Moreover, in some cases propeller
characteristics have to be optimized in correspondence to multiple
very different functioning points (i.e. different ship speeds, propeller
pitches...) including considerably off-design conditions, hardly
captured by conventional design methods, still widely based on lifting
line/surface approaches. Such designs, with a traditional approach,
would have been addressed in an intermediate condition, leading to a
geometry which is not optimal for any of the required settings.
Tools
• MatLAB for the parametric description of the propeller geometry,
• A Potential Panel Method, developed at the University of Genoa,
to efficiently compute propeller performances and steady/
unsteady sheet cavitation.
• modeFRONTIER, as a link between the parametric description
of the geometry and the panel method solver, to drive the
optimization through a multiobjective genetic algorithm (MOGA
II) for a total of 30.000 different geometries tested.
• StarCCM+ to further check the perforances of the original and of
the pareto geometries.
• The Cavitation Tunnel of the University of Genoa to finally validate
the results of the design by optimization through a dedicated
experimental campaign.
Application & Results
In this work, a «design by optimization», based on the coupling
between a multiobjective optimization algorithm and a panel code
(certainly more accurate and reliable with respect to traditional design
tools but inherently not directly applicable for the design itself), is
applied for the design of a Controllable Pitch (CP) propeller at different
pitch settings, with the aim of reducing the cavitating phenomena and,
consequently, the resultant radiated noise. Only through optimization,
as a matter of fact, it is possible to take advantage of the panel method
features in an «automated and iterative» procedure and to look for a
final design that correctly balance the performances at the different
working conditions on the basis of the objectives and of the constraints
required for the design itself.
Particular attention has been devoted to the slow speed (low pitch)
condition, obtained at constant RPM, and characterized by considerable
radiated noise and vibrations related to face cavitation. Numerical
results are validated by means of an experimental campaign, testing
both the original and the optimized geometry in terms of propeller
performances (delivered thrust and efficiency), cavitation extent and
radiated noise. Experimental results
confirm the numerical predictions, proving the capability of the
method to assess the propeller functioning characteristics and the
effectiveness of the proposed design procedure in correspondence of
challenging problems.
Objectives
Design a propeller that delivers the same thrust (both at the design
and at the reduced pitch working points) in order to:
• Reduce back cavitation at the design pitch.
• Reduce face cavitation at the reduced pitch.
• Avoid back cavitation at the reduced pitch.
• Avoid face cavitation at the design pitch.
• Increase the efficiency (both pitches).
CAE Poster Award
Prediction of the sheet cavity extension and of the propeller thrust and
torque by a computationally efficient Panel Method
The Pareto Frontier representation of the multi-objective optimization
carried out with modeFRONTIER
Newsletter EnginSoft Year 10 n°4 - 58
Thrust (KT), torque (KQ) and efficiency (h0 ) at the design pitch setting for the original and the optimal
propellers - Measurements at University of Genoa Cavitation Tunnel
Observed pressure side cavity extension for the original (left) and for the optimal (right) propeller at the reduced pitch setting for the
evaluation of the radiated noise - Validation of the design at the University of Genoa Cavitation Tunnel
Stefano Gaggero, Michele Viviani - University of Genova
59 - Newsletter EnginSoft Year 10 n°4
CAE Poster Award
CAE POSTER AWARD 2013: WINNER
Flow field and cavity extension prediction for the Pareto cases by a
RANS multiphase solver for the selection of the optimal geometry
CAE POSTER AWARD 2013: WINNER
CFD characterization and thrombogenicity analysis
of a prototypal polymeric aortic valve
The recent growth of cardiovascular diseases, primarily due to
heart valve failures such as stenosis and regurgitation, led to an
increasing need of valve replacement procedures.
Polymeric prosthetic valves seem to gain advantages over both
tissue and mechanical valves, as they combine excellent fluid
dynamic features with long lasting performance. Moreover,
they can be grafted via minimally invasive transcatheter aortic
valve implantation (TAVI), reducing related surgical risks and
complications.
Besides, heart valve optimization requires specific analysis
regarding the effects of blood contact. In particular, shearinduced platelet activation can be investigated in order to estimate
thrombongenicity.
This work presents a CFD approach for characterizing the
fluidynamics of a novel hemodynamically and functionally
optimized polymeric prosthesis for aortic valve replacement and to
evaluate its thrombogenic potential.
Materials and Methods
Geometry
Starting from the closed valve CAD model (Fig. 1a), a FEM
simulation was performed, on SIMULIA© Abaqus 6.10, with a
symmetrical boundary condition and an homogeneous loading
reproducing systolic transvalvular pressure. The systolic valve
configuration was therefore obtained (Fig. 1b).
The fluid domain (Fig. 2) was reconstructed on ANSYS© 13.0
Workbench Platform, in order to reproduce the
Valsalva Sinuses and the aortic arch with its three
upper branches.
Discrete Phase Modeling approach.
Particles loading history was
computed over 300 ms (time step 0.1
ms) and, via a statistical analysis, the
Stress Accumulation (SA) distribution
was extrapolated.
Results and discussion
The maximum velocity value, located
downstream of the valve housing,
was equal to 1.58 m/s, while blood
stagnation phenomena were detected within the Valsalva Sinuses.
The peak transvalvular pressure drop was equal to 2.41 mmHg.
Velocity magnitude contours are shown in Fig. 3, at the ejection
peak (T = 150 ms).
The valve prototype did not induce significant alterations into the
aortic fluid dynamic thanks to its morphological similarity with the
anatomy of the aortic valve. Pressure drop and maximum velocities
were comparable with those reported for other bioprosthetic valves.
Particle trajectories (Fig. 4a) were analyzed in order to quantify
the level of platelet activation due to shear stress. Scalar stress
and particle residence time were combined to calculate SA. The
statistical distribution of SA (Fig. 4b) allowed to identify particle
trajectories with a high thrombogenic risk.
CFD computational setup
Transient CFD simulations were run on ANSYS
Fluent v13.0. A systolic ejection waveform was
applied to the inlet section and outflow boundary
conditions (constant flow rates) were imposed to
the four outlets. A k-ω turbolence model with low
Reynolds corrections was adopted and blood was
characterized with density equal to 1060 kg*m-3
and viscosity equal to 3 cP.
Particle tracking
Neutrally buoyant spherical particles (Ф = 3 μm),
representing platelets, were injected into the fluid
domain from the inlet surface through a two-phase,
CAE Poster Award
Conclusions
The present study combined FEM and CFD
simulations to evaluate the hemodynamic
and thrombogenic performances of a novel
hemodynamically and functionally optimized
polymeric trileaflet valve.
Acknowledgements
The research leading to these results has
received fundings from the Cariplo Foundation
Project, Grant Agreement N° 2011-2241.
Figure 1 – CAD images of the closed (top)
and open (bottom) configurations
of the optimized polymeric valve
Filippo Piatti, Matteo Selmi, Alessandra Pelosi,
Alberto Redaelli - Politecnico di Milano
Thomas E. Claiborne, Danny Bluestein - Stony Brook
University, New York
Newsletter EnginSoft Year 10 n°4 - 60
Figure 3 – Contours of Velocity Magnitude [m/s] in the aortic arch and in
four different locations within the Valsalva Sinuses
Figure 4 – (a) Particle trajectories coloured by Scalar Stress [Pa]; (b) Frequency distribution of the Stress Accumulation [Pa·s] in the model
TECHNET MEETING
The TechNet Alliance Fall Meeting 2013 took place
at Lake Garda, Italy, with the support of EnginSoft
organization and was well attended by 73 members. The
official part of the meeting started on Friday morning
with an Oil & Gas Initiative Meeting. After lunch till late
afternoon, a meeting for ANSYS channel partners took
place, giving an updated overview of ANSYS roadmap
for the next future. In the evening, all attendees met for a
welcome dinner at the hotel. The main event, planned on
Saturday, started with the welcome of Stefano Odorizzi
to all attendees and a short informative presentation on
EnginSoft company and its field of activities. Afterwards,
Günter Müller from CADFEM gave an update on TechNet
Alliance and also the new webpage has been introduced
to the audience. Potential new members were invited to
give their presentations, on several different topics (from
aerospace to energy, from aeronautics to biomechanics,
among the others). On Saturday evening, the TechNet
Alliance Fall Meeting 2013 closed with a dinner for
all attendees at the excellent restaurant “Ai Beati” in
Garda.The next meeting, i. e. TechNet Alliance Spring
Meeting 2014, will take place in Malta on April 11th
and 12th, 2014.
61 - Newsletter EnginSoft Year 10 n°4
Technet Alliance
CAE is a complex and fast developing technology, it requires expertise in a variety
of disciplines. Service companies who can provide this expertise are typically small
to medium size enterprises- focused on one specific industry or discipline. No
single company exists that can possibly possess all of the world’s CAE knowledge
and experience. Therefore, it is difficult for a large company to find sufficient
CAE expertise to satisfy its needs. However, by combining the best engineering
talent, product knowledge, consulting expertise, training and support into a single
entity, an “Alliance of Experts” can collaborate to solve the most complex CAEproblems. Such an Alliance may function as a “virtual corporation” - providing a
high concentration of CAE expertise and services worldwide not available through
any individual company. the future, networking and building alliances becomes
vitally important to remain competitive in the global market. This is especially true
for small and medium sized companies. For these reasons a company, Technology
Network Alliance AG, was established 1998 in Switzerland by an international
group of CAE- service companies. By combining the unique expertise of many
companies into a “global corporation”, the Alliance is capable of focusing the
skills, resources, and people to fulfill a market need. Today. The TechNet Alliance is
perhaps the world’s largest network of engineering solution providers- dedicated
to the application, development, marketing and support of CAE software. Beyond
CAE service companies, business support companies, renowned professionals
from industry, professors from universities and even representatives of corporate
companies also belong to this network. www.technet-alliance.com
CAE Poster Award
CAE POSTER AWARD 2013: WINNER
Figure 2 – CAD 3D geometry of the flow domain within the aortic arch. The red
zone represents the Valsalva Sinuses where the prosthetic valve is placed
CAE POSTER AWARD 2013: WINNER
Thermo-Fluid Dynamics model of
two-phase system alloy-air inside
the shot sleeve in HPDC Process
In HPDC process, the final quality of castings is highly correlated to
the first stage of injection. During this phase, the movement of the melt
due to plunger’s acceleration causes the high level of air entrapment,
inducing porosity into the component. This has a detrimental effect
on mechanical properties and produces internal and surface defects.
To prevent these phenomena, it is important to control all the relevant
process parameters. In the present work, this objective has been
achieved through the development of a model that describes the
thermofluid dynamics behavior inside the shot sleeve. The model
consists in:
• Numerical model: implementation of thermal equation into an
open source CFD code;
• Mathematical model: generation and execution of a DOE. The
developed code has been used to simulate several cases with
different combinations of input parameters.
The models allow to determine the response surface that is used to
analyze the percentage of trapped air.
CAE Poster Award
The activity has been conducted for the master thesis work.
Numerical Model
Scope
Development of a numerical model that describes the dynamic of the
system inside the shot sleeve also considering the heat exchange.
Mathematical Model
Model Design
Scope
Identification of the relevant process parameters which will constitute
the input variables for DOE and parameterizing the model as a function
of these.
Process parameters selected
1. D = inner diameter of the shot sleeve;
2. L = length of the shot sleeve;
Newsletter EnginSoft Year 10 n°4 - 62
At the moment, the simulations are running and so the results are
partial.
Future aim
With this tool, it will be possible to explore all the combinations of
input parameters (within the given ranges) and to forecast trapped air.
This would support HPDC engineers in managing process parameters.
CAE POSTER AWARD 2013: WINNER
Roberto Meneghello, University of Padova
3. F = initial filling of melt.
4. V = velocity of the first phase;
Other consideration
All the remaining variables have been set as a function of four previous
input parameters. It has been necessary to define a reliable and
consistent condition for the end time of the simulations. The aim is
to properly compare the results of trapped air volume percentage for
different cases: simulations end when the total volume of the cylinder
equals the initial volume occupied by melt alloy.
Definition and execution of DOE
In this phase of project, modeFRONTIER has been used
as the basis for the DOE planning but not for DOE execution.
DOE definition
Each input variable range with a fixed quantization step. To avoid
unfeasible designs some constraints have been imposed on
combinations between input variables. In this way only the relevant
cases in foundry practice have been simulated. SOBOL algorithm has
been adopted to uniformly distribute a given number of experiments
in a design space.
DOE execution
It has been necessary to work out some scripts automatically
generating all the experiments by reading variables from a text file.
These cases have been executed on a cluster system which enables to
run several designs at a time using several processor.
Application of RSM
The rate of entrapped air (R) has been calculated by dividing the
final volume of air (Vair) respect the final total volume (Vtot). After
inserting these values in modeFRONTIER, it has been possible to
apply Response Surface Methodology (RSM) which correlates input
variables with the related output by means of a mathematical model.
63 - Newsletter EnginSoft Year 10 n°4
CAE Poster Award
CAE POSTER AWARD 2013: WINNER
FEM Analysis, Modelling and Control
of a Hexacopter
This work is supported by the PO. FESR 2007/2013 subprogram
4.1.1.1 “Actions to support the research and experimental
development in connection with the production sectors,
technological and production districts in areas of potentiality
excellence that test high integration between universities, research
centers, SMEs and large enterprises”; (Prog. “Mezzo aereo a
controllo remoto per il Rilevamento del Territorio - MARTE” Grant.
No. 10772131). The aim of the project is to realize a new platform
for the representation of the terrain in a georeferenced raster map by
using free and open source Geographic Information System (GIS).
In this wok a FEM structural analysis and the mathematical model
and control of a hexacopter airframe is presented. In particular, the
six-rotors are located on the vertices of a hexagon and they are
equidistant from the centre of gravity; moreover, the propulsion
system consists of three pairs of counter-rotating fixed-pitch
propellers in order to balance the torque actions. The structure has
been made up by a composite sandwich configuration characterized
by CFRP skins and closed cells foam core. The analysis has been
performed by the code ANSYS V 14.5 Academic Version with the
ACP tool to pre-postprocess the composite structures. In order
to describe the movement of the drone in space, an efficient
mathematical model has been introduced associated with a robust
control technique that has been implemented by means of MATLAB
software.
Mathematical Model and control
Supposed the drone as a rigid body, its dynamics is deduced from
the classical Newton- Euler equations but in terms of quaternions.
Taking into account all the internal and external influences, the
transational and rotational components of the motion read
in which m is the mass of the drone, ξ=(x,y,z) represents its
position vector with respect to the inertial frame, q=(q0,q1,q2,q3)
represents the quaternion describing the angular position, Fg is
the gravitational force, TB is the total thrust, Q is the orthogonal
transformation matrix from the body frame to the inertial one, S
is the velocity transformation matrix and n=(p,q,r) is the angular
velocity, I is diagonal inertial matrix, Γ represents the gyroscopic
effects and tB=(tf, tq, ty) the roll, pitch and yaw moment torque
vector. To maneuver the flight and to manage the hexacopter, a PID
control technique has been implemented.
A. Alaimo, V. Artale, C. Milazzo, A. Ricciardello, L. Trefiletti
University of Enna KORE
FEM structural analysis
• Units B are assembled on the mold (outer skin), according to
the B local reference system.
• Units A are assembled with a stacking sequence [O2/Core/
O2], where the angle 0° is coincident with xA axis. Then the
structure is completed by assembling the inner skin as well as
the outer one (unites B).
• A local annular reinforcement is stacked following the
[0/+302/0-302/0] layup referred to the C local coordinate
system.
The modal analysis of the hexacopter is performed with the aim to
compare the natural frequencies of the structure with the forcing
frequencies deriving from the thrust of the electric motors. The first
10 frequencies are listed in the table.
• Forcing frequencies from the thrust of the electric motors
(max rpm @ 9000)
• Hovering condition @1/2 Max power motors (4500 rpm)
• Normal flight regimes @ ± 20% Hovering rpm (3600 : 5400
rpm) → 60 : 90 Hz.n
So a superposition is possible from 7th to 9th mode.
CAE Poster Award
Figure 1 – View of the hexacopter assembled CAD geometry (Solid Works 2013)
Table: masses applied on the drone
Newsletter EnginSoft Year 10 n°4 - 64
Fig. 2 - Loads on the structure: weights in gravitational field considered as concentrated
masses; thrusts and reaction moments by electric motors
Fig. 3 - A, B, C are typical units of the layup with axial symmetry and last
picture shows the structure thickness
Fig 4. - Total deflection [mm] on the left, IRF under axisymmetric
load at n=3 on the right
Table 3 - First 10 frequencies
Fig. 5 - Modal shapes (from 7 to 9)
Fig. 6 - A,B,C are typical units of the layup with axial symmetry and last
picture shows the structure thickness
Fig. 7 - Rubik Cube: the symbol of the 2013 CAE Conference
65 - Newsletter EnginSoft Year 10 n°4
CAE Poster Award
CAE POSTER AWARD 2013: WINNER
Table 2 - Materials mechanical data
WEBINAR CFD e supporto fluidodinamico
L’offerta CFD di ANSYS copre una vasta gamma di applicazioni, attraverso l’utilizzo di svariati modelli fisici e numerici. Ad ogni nuova versione inoltre, tali modelli vengono migliorati in termini di efficienza e
stabilità ed estesi in termini di applicabilità. Nuovi modelli vengono
implementati, consentendo così agli utenti di affrontare problemi con
fisiche sempre più complesse.
Contemporaneamente a questo, i core solver traggono beneficio da
tutte le risorse che ANSYS impegna nello sviluppo delle soluzioni
HPC (High Performance Computing), attività strategicamente portata
avanti in costante collaborazione con i principali produttori hardware
(processori e schede grafiche) e consente ad ogni nuova versione
di essere apprezzabilmente più veloce della precedente, sia per il
solutore seriale che parallelo.
Ecco perché, ad ogni nuova versione il panorama di funzionalità offerto si amplia e nuove metodologie vengono rese disponibili, eventualmente corredate di esempi pratici.
Per tali ragioni, nel corso del 2014 EnginSoft ha pianificato con cadenza mensile, degli webinar tematici (legati alle novità software o
alla fisica), con lo scopo di fornire un aggiornamento più frequente e
strutturato ai propri clienti, in una modalità che sia la più fruibile possibile per la maggior parte degli utenti (un semplice collegamento
internet e per la durata di 50 minuti circa).
L’elenco di seguito riporta i webinar del primo Quarter.
• ANSYS Icepak: tool per la simulazione termo-fluidodinamica di
componenti elettronici
• ANSYS CFD Professional e modularità dei prodotto CFD di ANSYS
R15
• ANSYS CFX R15: novità del software
• Approccio multifisico: come realizzare analisi di interazione fluido-struttura (FSI)
• Multifase: stato dell’arte delle potenzialità multifase degli applicativi ANSYS CFD R15
• ANSYS HPC: potenzialità e offerta per il calcolo parallelo in ANSYS
• ANSYS Turbo System: novità di ANSYS R15 dedicate al mondo delle turbomacchine
• ANSYS Workbench R15 Parametric Workflow: peculiarità parame-
Events
•
•
triche dell’ambiente
Aeroacustica: stato dell’arte, modelli disponibili e relativi costi/
benefici
Combustione: stato dell’arte, modelli di reazione disponibili e loro
applicabilità
Ulteriori informazioni relativamente a ciascun webinar possono essere
trovate sul sito EnginSoft al seguente link:
http://www.enginsoft.it/webinar/index.html
Per iscriversi agli webinar, basta navigare all’interno dello stesso link e
formalizzare l’iscrizione. La partecipazione agli webinar è gratuita.
Un tecnico specializzato a disposizione per incontri
con i nostri clienti
http://www.enginsoft.it/rules/trules13.html
Dal 1° gennaio 2014 EnginSoft mette a disposizioni dei propri clienti in
regola con il contratto di manutenzione, un nuovo servizio di supporto
diretto nelle stesse date dei webinar. Tale servizio metterà a disposizione
degli utenti software un tecnico EnginSoft per 4 ore / anno (presso le
nostre sedi). In aggiunta, per i clienti ANSYS il cui contratto di manutenzione sia in corso di validità, è prevista, nelle stesse date dei webinar, la
possibilità di usufruire di una mezza giornata di affiancamento dedicata,
a valle o a monte dell’webinar. Ciò avvalendosi di uno dei tecnici EnginSoft, che seguirà l’utente, affiancandolo sulla tematiche di suo interesse
o su caso specifico che l’utente potrà portare con se nell’occasione.
La mezza giornata è da intendersi come affiancamento dedicato su tematiche specifiche del cliente, il cui intento è di condividere con lo stesso una metodologia e/o fornire qualche indicazione specifica a riguardo
di un progetto.
All’atto dell’iscrizione alla mezza giornata, l’utente avrà tuttavia la possibilità di spiegare, attraverso qualche riga di testo, la tematica oggetto di
discussione, in modo da avere a disposizione la risorsa di EnginSoft più
adeguata al proprio tema.
La mezza giornata verrà svolta presso la sede EnginSoft di Bergamo
(http://www.enginsoft.it/dove/bergamo.html) ed è aperta ad un numero
massimo di 10 utenti per sessione.
Newsletter EnginSoft Year 10 n°4 - 66
EVENT CALENDAR
June 11-13, 2014
Verona, Italy
METEF 2014
http://www.metef.com
Expo of customized technology for the aluminium&innovative metals
industry. EnginSoft will be present with a booth.
October, 2014
International CAE Conference
http://www.caeconference.com
EnginSoft will be the main sponsor of the International CAE
Conference. Many of our engineers will be engaged in presenting
industrial case histories during the parallel sessions of the
Conference or technologies update during the workshops.
CAE WEBINARS 2014
EnginSoft continues the proposal of the CAE Webinar on
specific topics of the virtual prototyping technologies,
such as: non linear phenomenas, turbomachinery,
meshing, parametric workflow, optimization…
The CAE Webinar program will
grow up during 2014 with many
other topics on simulation.
Stay tuned to www.enginsoft.it/webinar
for the complete program of webinars.
The podcasts on past CAE Webinars
are available at:
www.enginsoft.it/webinar
2014 CAE EVENTS
Stay tuned to www.enginsoft.it/eventi
for the complete program of the events in 2014
67 - Newsletter EnginSoft Year 10 n°4
EnginSoft at SPE Offshore Europe 2013
The biannual SPE Offshore Europe Conference held in Aberdeen was
attended internationally by over 63,000 people and 1,500 exhibitors
across the 4 days from the 3rd-6th September.
The SPE Offshore Europe Conference is one of the largest Oil &
Gas events attracting a global audience that were keen to explore
leading innovation technologies and services available. “The Next 50
Years” theme gave a fantastic platform of opportunity for EnginSoft to
understand the ongoing and predicted challenges that will be faced in
the Oil & Gas industry and establish their presence as specialists in
complex simulation to an international audience.
With a 3 year waiting list for stand space, EnginSoft exhibited, with
thanks, at Offshore Europe as part of a stand share with NAMTEC
(National Metals Technology Centre). The conference provided an
opportunity to meet with existing customers and proved to be a great
event to showcase EnginSoft’s expertise in complex simulation with the
support of Stefano Odorizzi, Massimo Galbiati, and Livio Furlan.
EnginSoft for the Oil & Gas industry
EnginSoft operate as a key partner in Design Process Innovation, we
specialise in Complex Simulation and Optimisation activities, including:
• Offshore FEA/CFD
• Oil & Gas Equipment
• Flow Modelling
• Subsea & Geology
• Reservoir
For more information: [email protected]
Your FREE Newsletter Copy
Do you want to receive a free printed
copy of the EnginSoft Newsletter?
Just scan the QRcode or connect to:
www.enginsoft.it/nl
Events
21 | 22 OCTOBER 2013
Pacengo del Garda
Verona - Italy
THANKS TO THE 1000 PARTICIPANTS
AND SEE YOU IN 2014!
CAE CONFERENCE PROCEEDINGS ARE
AVAILABLE TO DONWNLOAD ON:
http://proceedings2013.caeconference.com
www.caeconference.com
Author
Document
Category
Uncategorized
Views
0
File Size
5 220 KB
Tags
1/--pages
Report inappropriate content