Transactions of Nanjing University of Aeronautics and Astronautics
Modeling methods and Test verification of the root insert contact
interface for wind turbine blade
Li Hui,Wang Tongguang*
(Jiangsu Key Laboratory of Hi-Tech Research for Wind Turbine Design, Nanjing University of
Aeronautics and Astronautics, Nanjing, 210016, China)
ABSTRACT: Two modeling methods of the root insert for wind turbine blade are presented in this paper, which are local
mesh optimization method (LMOM) and global modeling method (GMM). Based on the optimized mesh of the local
model for the metal contact interface, LMOM is proposed to analyze the load path and stress distribution characteristics.
While GMM is used to calculate and analyze the stress distribution characteristics of the resin which is established
between the bushing and composite layers of root insert. To validate the GMM, the tension test is carried out. The result
successfully shows the shear strain expresses a similar strain distribution tendency with the GMM’s results.
KEYWORDS：root insert; modeling methods; mesh optimization; contact interface; tension test
Foundation item: The National Basic Research Program of China ( 973 Program) ( 2014CB046200)
related to the configuration, design and process
0 Introduction
manufacturing method. In 1990, the connection of root
Currently, the connections of the wind turbine
insert was firstly published by LM which created the
blade root mainly adopt T-bolt and root insert. T-bolt
precedent of the root insert configuration design[3]. In
modeling and analysis method has been mastered by
2009, GAMESA proposed the bushing and composite
many domestic research institutions. Li Dong-hai uses
layers of root insert had two forms of double shear
ANSYS to establish the finite element analysis model
joint[4-5]. In 2010, Vestas claimed that the length of
[1]
of T-bolt , the stress analysis and strength check are
bushing design could be changeable[6]. In 2014,
carried out for FRP and bolts of root connection in
LianZhong proposed a method to manufacture root
order to guide the design, optimization and material
insert bushings for wind turbine blade by pultrusion
selection for T-bolt; PAN Yan-huan and his group focus
process[7].
their research on the method for FE modeling of
[2]
In recent years, with the rapid development of
complicated layered structures . In consideration of
large wind turbines, there is a growing trend of large
the characteristics of repeated layers in root structure
blades. T-bolt due to its own limitations is gradually
for wind turbine blade, the sub-laminate method was
replaced by root insert. However, there are few
introduced for calculating equivalent stiffness, the FE
publicly available resources related to the technical
model of root structure was established while it was
research of root insert[8]. Therefore, it’s the key
treated as unique material, the effect of loads on
technology to master the accurate modeling and
boundary conditions and contact states between nuts
calculate analysis for root insert of blade design. There
and holes was also discussed.
is a possibility that the efficient and accurate modeling
However, root insert technique is still in the
and simulation for root insert may provide a more
product design and test verification phase for domestic
accurate simulation data for the strength test which
wind power research institutes so far. The key
needs long circle and high price, or even replace it.
analytical methods are monopolized by the minorities
of foreign research institutions who only public patents
*Corresponding author:Wang Tongguang, Professor, E-mail：[email protected]
Transactions of Nanjing University of Aeronautics and Astronautics
 fr  u nt
1 Coulomb Friction Contact
According to the root insert of wind turbine blade,
Where  n and
the nonlinear contact of the contact interfaces are
analysed using the coulomb friction model. Mechanics
define contact as a complex problem, because the
boundary
conditions
are
highly
nonlinear.
The
(2)
 fr denote the normal stress and
the tangential friction stress of contact node,
respectively.
u is friction coefficient, and t is the
movement of the objects should be traced accurately,
tangential unit vector in the direction of the relative
and the interactions of those should be traced as well
slipping velocity.
after the contact. The two contact objects must fit the
Coulomb friction is a highly nonlinear property
no penetration constraint condition, which fit the
that relies on the normal force and the relative sliding
formula (1):
speed, it is an implicitly function of velocity and
u A * n  D
(1)
Where u A is the incremental displacement vector
of point A , n is the unit normal vector, and D is the
contact distance tolerance.
displacement increment. The value consists two parts:
one is the contribution of the tangential friction force,
and the other is the contribution of the system stiffness
matrix.
Coulomb friction is very widely used, Coulomb
friction contact model expressed in formula (2):
2 The Mesh Optimization and Analysis of
the Metal Contact Interface
widely used in wind turbine engineering at present.
Figure1 shows the integral component type and the
model relationship.
Modeling analysis and test model of this paper are
selected from teeth-bushing and cylinder bolt which are
Figure 1 Integral component type and the model relationship of root insert
2.1 Optimization of the mesh topology
contact interface, which can be very effective to avoid
the stress concentration caused by the mesh quality,
In order to discuss the load path of the bushing
and the characteristics of the contact stress, 2D
and improved the nonlinear accuracy of the contact
area.
cross-sectional symmetrical model is used because the
To simplify the structure reasonably, high order
components of the root insert are complicated. The
accurate topology optimization with 1 to 2 mesh is
model is divided into three main metal member which
applied to contact areas manually (Figure2), and a few
called bolt, bushing and flange. Mesh topology
automatic meshes acted on non-power transmission
optimization model is built for the main load path of
Transactions of Nanjing University of Aeronautics and Astronautics
parts. Three load cases are carried out on the 2D model
and set the coefficient of friction as 0.2, the
which are preload, preload and tension, preload and
characteristics of the shear stress distribution in the
compress, using the Coulomb friction contact model,
metal contact interface are calculated.
Bolt
Bushing
Flange
(a) Flange-bolt contact interface
(b)
Flange-bushing contact interface
(c)
Thread contact interface
Figure 2 Topology optimization of metal contact interface
0.2
Analysis of the metal contact interface
1、Flange-bushing contact stress analysis
The distribution stress curves of the Cauchy stress
is consistent with the stress contour. The external force
and the contact normal stress in the flange and bushing
is transferred from the flange-bushing contact surface
contact interface elements are calculated in three load
through bushing, so the tension unloads the bushing,
conditions. The abscissa is shown in Figure2, the
while the compress uploads the bushing. The force
correspondence of the flange-bushing contact elements
transferring path between the flange and bushing is
are from top to bottom. Figure3 shows the distribution
verified by the three Cauchy stress curves.
trends of Cauchy stress in the contact interface area
increases from the inside out, stress distribution curve
Figure 3 Cauchy stress curve (left) and stress contour (right) of contact interface between flange and bushing
Transactions of Nanjing University of Aeronautics and Astronautics
2、Flange-bolt contact stress analysis
Figure4 shows the distribution stress curve of the
flange-bolt contact interface elements. The abscissa is
shown
in
Figure2,
the
correspondence
of
the
flange-bolt contact elements are from top to bottom.
The
external
force
is
transferred
from
the
flange-bushing contact surface through bushing, the
flange-bolt contact interface stress level of the three
load conditions are substantially the same. Stress
concentration is appeared in the corner, the local stress
Figure 4 Cauchy stress curve of contact interface between
flange and bolt
exceeds the material yield strength, therefore the
material is set to elastic-plastic and recalculated.
Figure5 shows the stress distribution of the flange-bolt
contact section, which explains that the elastoplastic
modeling can improve the stress concentration in the
corner to a certain extent after using the characteristics
of the elastoplastic material. At this point, the contact
element generates plastic deformation, and the contact
Cauchy stress and normal stress level are within the
material yield strength range.
Figure 5 Normal stress distribution of contact interface
between elastoplastic flange and elastoplastic bolt
3、Thread analysis
Figure6 shows the normal stress distribution of the
threads. The abscissa is shown in Figure2, the
correspondence of the thread contact elements are from
left to right. Figure7 shows the normal stress
distribution results of the first five threads, the stress
distribution of the thread contact elements was greatly
influenced by the contact analysis of the thread section,
which presents obvious fluctuation. The first thread
stress is the largest, the rest is in descending order, and
Figure 7 Normal stress curve of threads contact interface
the fifth thread has fallen about 50%, so it’s very
Above all, mesh topology of the metal contact
important to focus on the load capacity of the first 4-5
interface for root insert is optimized, and the analysis
thread in calculating and analysing.
accuracy of the contact interface is improved. In the
case of uploading the preload and external force on the
flange-bushing, flange-bolt and thread, using the
Coulomb friction contact model, the load path of root
insert is given, the distribution characteristics of the
Cauchy stress and normal stress in the metal contact
interface are also calculated. Based on the nonlinear
contact analysis, while the interface is optimized by
mesh topology, the stress distribution is still not strictly
Figure 6 Normal stress contour of threads contact interface
continuous, this may be associated with the nonlinear
force of the contact surface.
Transactions of Nanjing University of Aeronautics and Astronautics
3 FE Modeling and Shear Stress Analysis
of the Contact Interface between Bushing
and Composite Layers
multiple complex contact constraint for the entire root
insert, the relationships of the constraint components
are shown in table 1. There are eleven pares of
constraints, using the Coulomb friction model and the
Chapter 2 mainly discusses the mesh optimization
friction coefficient is 0.2.
modeling and contact analysis of the metal contact
interface for root insert, while this chapter sets the
Table 1 Multiple complex contact constraints of integral root insert
ID
Contact components
1
Plug
Busing
2
Resin
3
Composite layers/Bushing/Resin/UD circular/UD triangle/UD wedge
Flange
4
Flange
Bearing inner ring
5
Blade nut
Bearing inner ring
6
Hub nut
Bearing outer ring
7
8
Hub
Bearing inner ring
Bearing ball top
9
10
Bearing outer ring
Bearing inner ring
Bearing ball bottom
11
Bearing outer ring
To calculate the interfacial shear stress between
ratio is 0.3, see Figure9, using Coulomb friction model,
bushing and composite layers, considering the form of
the contact is analysed with both preload and tension
manufacturing process, an isotropy resin layer is
loading at the same time. After the simulation, the
established on the surface of the composite layers, the
in-plane shear stress of resin layer is extracted
thickness of the layer is 0.1mm, while the elastic
(Figure10) which appears the stress is distributed
modulus of the resin layer is 3000MPa and the poisson
evenly.
Frictional contact of
bushing and resin
Figure 8 Resin modeling between bushing and composite
layers
Figure 9 In-plane shear stress contour of resin layer
Along the spanwise direction of blade root, the
by the teeth shape desigh of the bolt. The strength
In-plane shear stress which in the middle of the resin
analysis assumption of the connection part is
level is extracted, the distribution of the values are
mentioned by the universal wind turbine design
shown in Figure11. The distribution of the in-plane
standard GL2010[9]. It believes that, both in the blade
shear stress along the spanwise direction is tended to
web-shell joints and the blade leading-trailing edge
balance, a certain volatility is existed which is caused
joints, the shear stress distribution of joints could be
Transactions of Nanjing University of Aeronautics and Astronautics
seemed as a steady shear stress distribution. The root
the analysis stresses of which are not completely steady
contact is not mentioned and the contact interface of
state distribution.
the root insert is also not involved in GL2010 because
Figure 10 In-plane shear stress curve of resin layer
4 Tension Test of the Root Insert
An experiment of root insert integral component is
measured in the test sections at six locations by the use
shown in Figure12, the strain gauges are embedded
of MTS to load tension force, see Figure13.The test
equidistantly along the longitudinal direction in the
begins with loading started and ended when the bolt is
contact interface between bushing and composite layers,
pulled out. During the test the force-time curve of root
and the distance of each is 50mm. Local strain are
insert is monitored, the strain is also recorded.
Figure 11 Manufactured products of root insert
Figure 12 Tension test of root insert
The two-way tensile loading is used in the
Therefore, the method of modeling and analysis by
experiment. The calculation model of the root inset and
FEM seems to be very promising and can be used in
the analysis method are discussed in chapter 3. In
root insert design for wind turbine blade. Nevertheless,
Figure14, a plot shows the strain distribution trends of
there is a certain difference of the stress distribution
the experimental result when the bolt is pulled out and
near the loading section between the experimental
the calculation result by GMM’s FE model. It is proved
value and calculated value because of the influence by
to be a similar strain distribution to a certain extent.
loading boundary condition.
Transactions of Nanjing University of Aeronautics and Astronautics
Figure 14 Strain distribution of interface between bushing and composite layers
3
Connection for Wind Turbine Blade[J], FRP,2010
Conclusion
(4):1-5.
LMOM and GMM are successfully simulated and
[2]. PAN Yan-huan, Finite Element Modeling of Wind
tested to analyze the root insert interface for wind
Turbine Blade Root[J], JOURNAL OF NORTH
turbine blade according to the Coulomb friction contact
UNIVERSITYOF
model. LMOM analyzes the load path of the root insert
EDITION), 2010.31(4):429-432.
and the stress distribution characteristics of the metal
[3]. John E.Eckland;James V.Frerotte, Wind turbine
contact
blade
CHINA(NATURAL
methods[P],
US4915590,
characteristics of the isotropous resin between the
[4].Ion AROCENA DE LA RUA,
Eneko Sanz
bushing and composite layers on the premise of the
Pascual,
multiple complex contact constraints. The GMM
US2009/0324417A1,Dec.31,2009.
results are tested by the tension test of the root insert
[5]. Ion AROCENA DE LA RUA, Eneko Sanz Pascual,
integral component which shows a similar strain
et
distribution tendency. This provides a good modeling
Dec.31,2009.
and analysis basis for the calculation and simulation of
[6].
the root insert for wind turbine blade. LMOM and
turbine
GMM provide a good modeling and analysis basis for
May14,2010.
the calculation and simulation of the root insert for
[7]. Huang Huixiu,Wang Zhiwei,el,at. A method to
wind turbine blade. Although a lot of improvement can
manufacture root insert bushings for wind turbine blade
be made in LMOM and GMM, we see a possibility that
by pultrusion process[P],CN103909662A,July.9,2014.
those models can be developed to give accurate
[8]. Sa Hao-liang, Strength Analysis Method and
analysis predictions to make root insert test become
Experimental Study on Root Bolt of Wind Blade in
unnecessary.
Pre-burying Bolt Sleeves[D], WuHan Polytechnic
interface
calculates
and
by mesh optimization;
analyzes
the
stress
GMM
distribution
attachment
SCIENCE
Apr.10,1999.
et
al.
al.
Blade
Blade
insert[P],US2009/0324420A1,
VRONSKY,Tomas,HAHN,Frank,et,al.
rotor
insert[P],
blade[P],
Wind
WO/2010/052487A2,
university,201306.
Reference
[9]. GL2010 IV-Part 1, Guideline for the Certification
of Wind Turbines[S].
[1]. Li Dong-hai, Finite Element Analysis of the Root