Elsevier Editorial System(tm) for Energy Conversion and Management
Manuscript Draft
Manuscript Number: ECM-D-14-01362R1
Title: Predictive tool of energy performance of cold storage in agrifood industries: The portuguese case
study
Article Type: Original research paper
Section/Category: 1. Energy Conservation and Efficient Utilization
Keywords: Energy efficiency; sustainability; perishable products; cold storage; predictive tool; cold
industry.
Corresponding Author: Prof. Pedro D Gaspar, Ph.D.
Corresponding Author's Institution: University of Beira Interior
First Author: José Nunes, PhD
Order of Authors: José Nunes, PhD; Diogo P Neves, MSc; Pedro D Gaspar, Ph.D.; Pedro D Silva, PhD; Luís
P Andrade, PhD
Abstract: Food processing and conservation represent decisive factors for the sustainability of the
planet given the significant growth of the world population in the last decades. Therefore, the cooling
process during the manufacture and/or storage of food products has been subject of study and
improvement in order to ensure the food supply with good quality and safety. A predictive tool for the
assessment of the energy performance in agrifood industries that use cold storage is developed in
order to contribute to the improvement of their energy efficiency. The predictive tool for analysis of
the energy performance is based on a set of characteristic parameters from mathematical correlations:
amount of raw material, annual energy consumption and volume of cold rooms. Case studies of
application of the predictive tool consider industries in the meat sector, specifically slaughterhouses.
The results obtained help on the decision-making of practice measures for improvement of the energy
efficiency in this industry.
Cover letter
Professor Moh'd Ahmad Al-Nimr
Jordan University of Science and Technology, Irbid, Jordan
Enclosed is the revised paper that was rigorously verified and modified to fully address
the comments and suggestions made by the reviewers. Additionally, as suggested by
you, the literature survey was updated.
We include three files:
(1) A "Response to Reviewers.doc" file, containing a reply to the reviewers comments
and suggestions.
(2) A word document named "Manuscript_Total_Highlight.doc" containing the revised
text with coloured highlighting.
(3) A word document named "Manuscript_Total.doc" containing the revised manuscript
without any annotations, highlighting or comments.
With the modifications and additional data included in the manuscript based on the
reviewers and editor comments and suggestions, we are now confident that the revised
manuscript is worthy of publication in Energy Conversion and Management Journal.
Yours Sincerely,
Pedro Dinis Gaspar
Highlights (for review)
Highlights

A predictive tool for assessment of the energy performance in agrifood industries
that use cold storage is developed.

The correlations used by the predictive tool result from the greatest number of data
sets collected to date in Portugal.

Strong relationships between raw material, energy consumption and volume of cold
stores were established.

Case studies were analysed that demonstrate the applicability of the tool.

The tool results are useful in the decision-making process of practice measures for
the improvement of energy efficiency.
*Complete Manuscript including All Figs & Tables
Click here to view linked References
1
Predictive tool of energy performance of cold storage in agrifood
2
industries: The portuguese case study
3
4
José Nunes1, Diogo Neves2, Pedro D. Gaspar2,*, Pedro D. Silva2, and Luís P. Andrade1
1
5
Politechnic Institute of Castelo Branco, Av. Pedro Álvares Cabral nº12 6000 Castelo
6
Branco, Portugal
7
[email protected], [email protected]
8
2
University of Beira Interior, Electromechanical Engineering Department, Rua Marquês
9
d’Ávila e Bolama 6200-001 Covilhã, Portugal
10
[email protected], [email protected] (* corresponding author), [email protected]
11
12
Abstract
13
Food processing and conservation represent decisive factors for the sustainability of
14
the planet given the significant growth of the world population in the last decades.
15
Therefore, the cooling process during the manufacture and/or storage of food products
16
has been subject of study and improvement in order to ensure the food supply with good
17
quality and safety. A predictive tool for assessment of the energy performance in
18
agrifood industries that use cold storage is developed in order to contribute to the
19
improvement of the energy efficiency of this industry. The predictive tool is based on a
20
set of characteristic correlated parameters: amount of raw material annually processed,
21
annual energy consumption and volume of cold rooms. Case studies of application of
22
the predictive tool consider industries in the meat sector, specifically slaughterhouses.
23
The results obtained help on the decision-making of practice measures for improvement
24
of the energy efficiency in this industry.
25
26
27
Keywords: Energy efficiency; sustainability; perishable products; cold storage;
predictive tool; cold industry.
28
29
1
Introduction
30
The sustainability and food safety in the agrifood sector has been a topic of research
31
and study. In the early twentieth century, the world population was about 1500 million
32
people, however just over a century the number of habitants increased to approximately
33
7000 million [1]. Thus, it is clear that the world demand for food is increasing which is
34
reflected on an enormous strain for the production and conservation sectors of the food
35
chain. So it becomes imperative to find short-term solutions for the agrifood industries
36
that in a macroscopic vision will help the sustainability of the planet. In this context,
37
cooling plays an important role, allowing to store food products in times of increased
38
production when the market has no capacity for product flow, or just make them
39
available when needed. The refrigeration process has the ability to preserve perishable
40
products ensuring their chemical, physical and nutritional properties but it is also
41
indispensable in the processing of perishable foods such as meat, fish, milk, fruit and
42
other. The demand of chilled and frozen food as increased substantially, which leads to
43
new requirements for the efficiency of refrigeration systems [2]. Energy represents a
44
factor of greatest importance not only for a country economy, but especially for the
45
well-being of its citizens. In 1971 there was a global electricity consumption of around
46
439 Mtone, and in 2010 this figure rose to 1.536 Mtone [3]. This increase is related to
47
the population growth, being the industry responsible for about 35.2% of energy
48
consumption worldwide in 2010. The energy consumed globally for the cooling process
49
accounts for 15% of energy consumption [1].
50
Thus, it is important to develop studies and tools that help the efficiency
51
improvement of these processes to ensure a better sustainability of this industrial sector.
52
There are some results provided by studies and projects in this area, more specifically in
53
the development of predictive tools for the analysis of several points related to the
54
cooling processes (whether energy consumption, environmental impact, among others).
55
The tool FRISBEE CCP (Cold Chain Predictor) [4-6], developed within the activities
56
of FRISBEE (Food Refrigeration Innovations for Safety, Consumers Benefit,
57
Environmental impact and Energy) project, performs simulations on specific conditions
58
defined by the user, constructs graphs representing the variation of temperature over
59
time and predicts the remaining shelf time of the product. These simulations are
60
performed based on the method of Monte Carlo [4] generating distributions of
61
time/temperature for every stage of the cold-chain and selected product. The results
62
represent realistic scenarios for the behaviour of food products, being possible to take
63
corrective actions in order to optimize the efficiency of the cold-chain ensuring product
64
quality and increasing its validity. This tool is intended to provide free information and
65
tools to managers, designers and operators of the refrigeration industry optimizing the
66
efficiency of their companies. In this same context, Foster et al. [7] describes the ICE-E
67
(Improving Cold Storage Equipment in Europe) project [8] devoted to the development
68
of tools with the same goals. This project contributed to reduce the energy consumption
69
and greenhouse gas emissions from the European food cold storage sector through the
70
application of more energy efficient equipment and taking into account the energy and
71
environmental standards of the EU [8-10].
72
Eden et al. [11] describes a predictive tool, QMRA (Quantitive Microbial Risk
73
Assessment) [12-13], which is combined with the principle of the Hazard Analysis and
74
Critical Control Points (HACCP) to enable the improvement of quality and safety of
75
food in a preventive approach. More specifically, this tool allows to estimate the risk of
76
pathogens growth based on the temperature and chemical and nutritional characteristics
77
of food. As result, the SLP (Shelf Life Predictor) [13] was developed with the ability to
78
predict the true and the remaining shelf life of foods. Apart from the quality and safety
79
of foods, this tool allows tracking (traceability) products in order to know the places
80
where they are in real time so that the consumer can make use of reliable information
81
about the origin of food. It also allows to overlook the manufacturing process from the
82
collection of raw materials to their processing, helping to ensure quality and safety.
83
Van der Sluis [14] describes a software that makes the temperature control of cold
84
stores taking into account the price of electricity and the daily consumption profile of
85
the company. The load management was accomplished considering that compressors
86
will work during time periods of low-cost electricity due to lower demand, in order to
87
accumulate energy in the form of cold to rationally use it during peak hours [15].
88
However, this procedure is executable only if the facilities have enough cold storage
89
capacity to store the energy demand during the day. Besides the abovementioned works,
90
other computational tools and models have been developed to enhance the operation of
91
refrigeration facilities [16] and system devices, such as compressors and their control,
92
regulation and command [17-18], expanders [19], among others. Additionally,
93
methodologies, strategies and procedures has been developed for optimal control of heat
94
recovery processes in refrigeration systems [16], for the performance evaluation of heat
95
exchangers [20] as well as knowledge based decision making methods for refrigerant
96
selection [21] and energy savings based in energy audits and its detailed analysis [22].
97
These works are only few examples of recent ongoing research and development
98
focusing the energy improvement of refrigeration systems, that are the main consumers
99
in the agrifood industry.
100
This paper presents a predictive tool developed for the assessment of the energy
101
performance in agrifood industries that use cold storage. The tool uses mathematical
102
correlations from the greatest number of data sets collected to date in Portugal.
103
Statistical correlations between raw material, energy consumption and volume of cold
104
rooms are used. To show the features and applicability of the tool, several case studies
105
for different subsectors of the meat industry are analysed. The tool results will be useful
106
in the decision-making process of practice measures for the improvement of energy
107
efficiency.
108
109
2
Material and Methods
110
2.1
Framework in the context of Portuguese industries
111
The abovementioned research covers the development of predictive tools aimed to
112
improve the energy efficiency of cold stores since refrigeration is responsible for about
113
60-70% of electricity consumption of these facilities [23]. Although, Portuguese
114
companies had not been included in the databases. In this context, a project was
115
developed in Portugal focussed in the identification of energy consumption profiles in
116
the agrifood industry. Additionally, it intends to promote and develop actions that
117
contribute to a real improvement of energy efficiency and competitiveness of this
118
sector. The work developed and presented in this paper is part of a project activity and
119
its main objective is the development and implementation of an analysis algorithm to be
120
validated with companies outside the sample of technical audits. As part of the activity,
121
it is aimed the development of a predictive tool to support strategic decisions in
122
companies allowing the estimation of their energy performance and pointing practice
123
measure that lead to an effective improvement of energy efficiency. Note that this
124
project is not aimed for the energy characterization of the cold stores in particular, but to
125
the energy characterization by agrifood category, i.e., meat, fish, dairy, horticultural
126
(fruit and vegetables), wine and vineyard and distribution, in order to obtain real data
127
that was used for the development of the correlations included in the model/algorithm.
128
The Portuguese industry is mainly constitute by small and medium enterprises. In
129
2011, it number was around 1.112.000, providing employment to about 78.5% of the
130
Portuguese population [24-25]. A large part of these industries are included in the
131
agrifood sector.
132
133
2.2
Sample
134
This study was conducted based on a sample of 87 agrifood companies: 33 from the
135
meat sector, 31 from the dairy sector and 23 from the horticultural (fruits and
136
vegetables) sector in the central region of Portugal. According to INE [26], this region
137
has the highest number of agrifood companies in Portugal. Additionally, this region has
138
the perfect weather conditions and soil and vegetation properties that allows the
139
development of agrifood practices for a wide range of products from different sectors.
140
Several regional products are manufactured in the region, among which stand out the
141
products in the meat sector (cured meats), dairy products (cheeses) and fruits.
142
In particular, the meat sector industries, where the predictive tool was applied and
143
validated, were subdivided into three subcategories: Slaughterhouse, Sausages industry
144
and Hams industry. The analysis was conducted in 4 Slaughterhouses plants, 20
145
Sausages industries and 9 Hams industries. 70% of the ham manufacturing companies
146
are located in this region of the country.
147
148
2.3
Data Collection
149
Initially, an extensive survey was developed to carry out a collection of information
150
about the agrifood industries. This survey was used to collect information regarding
151
physical characteristics of plants, activities and characteristics of production processes,
152
high energy-use equipment with special emphasis on cooling systems, cooling and
153
freezing chambers and energy consumption. Besides serving as a support document for
154
data recording, the document ensured a systematic and reliable data collection.
155
Moreover, it was also used as a technical guide to the plants inspection. The inquiry
156
document was used to record the following data:
157
i) Identification and location of agrifood plants;
158
ii) Characterization of infrastructure (location, age, type of materials, dimensions and
159
160
161
location of cooling and freezing chambers);
iii) Characterization of activities (type and quantities of raw materials and final
products);
162
iv) Identification and characterization of the production process;
163
v) Identification and quantification of energy types;
164
vi) Characterization of the electricity tariff and consumption (tariff type, electrical
165
power contracted, electricity consumption, power factor);
166
vii) Breakdown of electricity consumption by sector and equipment;
167
viii) Characterization of cooling and freezing chambers (number, type of insulation
168
169
170
material, preservation condition, raw materials/products load);
ix) Determination of internal and external environmental conditions, of facilities and
cooling chambers (air temperature and relative humidity);
171
x) Identification and characterization of refrigeration systems (age, location, type of
172
technology and technical characteristics, refrigerant type and nominal electrical powers
173
of compressors);
174
xi) Evaluation of the characteristics of compressed air systems;
175
xii) Characterization of thermal fluid generators;
176
xiii) Evaluation of energy efficiency improvements.
177
The collected information relates to activities undertaken during 2008.
178
The data collection process was carried out face-to-face, with a visit to the industries,
179
and where the building, the equipment, the technical operations and the manufacturing
180
processes were observed. Data collection was performed during the year 2009.
181
182
2.4
Equipment and measurement techniques
183
In order to measure the environmental conditions inside and outside the cooling
184
equipment, including air temperature a006Ed moisture content, a digital multi-function
185
measuring equipment (Testo 435-2) was used, along with its temperature and humidity
186
probes (±0.3°C accuracy). The previous equipment with a contact probe with type K
187
thermocouple (±0.5°C accuracy) was used to measure the walls/surfaces temperature.
188
Temperature data-loggers (Microlite and Lascar electronics with (±0.5°C accuracy)
189
were used to measure the air temperature inside the cooling and freezing chambers. A
190
thermographic camera Testo 880 (±2°C accuracy) was used to evaluate the preservation
191
condition of cooling chambers walls, thermal bridges, sealing condition of chambers
192
doors, and insulation of refrigerant suction pipes and of discharge air ducts of air
193
handling units. An ammeter clamp was used for measuring the electrical current and
194
voltage input to the compressor motor, in order to evaluate its electrical power
195
consumption. The electrical power data obtained experimentally was compared with the
196
estimated values available from official catalogues and dedicated software, for the same
197
operating conditions. Close results (deviations below 10%) were found between the two
198
methods in use. Although the energy intakes were quantified by analysis of the monthly
199
electricity bills, a profile analysis of electricity consumption was also performed using
200
an energy analyser (Elcontrol Energy Explorer).
201
202
203
The physical dimensions of the cooling and freezing chambers were calculated using
an infra-red distance meter Bosch-DLE 40 (± 1.5 mm accuracy).
204
2.5
Database
205
A computer database was built to record all information gathered from the plants.
206
This database was designed to allow data processing and analysis. It also made possible
207
to obtain various performance indicators from each agrifood company and
208
simultaneously to carry out a comparative analysis between different plants in each
209
sector. Additional details of the database can be found in [27].
210
211
3
Predictive tool features
212
3.1
Overview
213
The predictive tool, Cool-OP (Cooling Optimization Program), was developed in
214
MATLAB [28] making use of GUIDE (Graphical User Interface Design Environment)
215
[29]. This tool provides point-and-click control of software applications, reducing the
216
need to type commands in order to run an application. GUIDE provides tools for
217
graphically design the user interface for applications. The graphical user interface
218
typically contains controls named components such as menus, toolbars, radio and push
219
buttons, edit text, static text and check boxes, sliders, among others. Then, GUIDE
220
automatically generates the MATLAB code for the user interface, which was
221
programmed with the behaviour expected for the application. The computations to be
222
performed by the application were programmed in MATLAB language and the data is
223
display in plots.
224
The user interface of the predictive tool is extremely simple, intuitive, user-friendly
225
and easy to understand. The aim is to provide a tool accessible to all professionals in the
226
agrifood industry regardless of their qualifications. Finally, the code generated in
227
MATLAB was converted to an executable file that can run in any computer.
228
229
230
231
3.2
Algorithms
The predictive tool provides the user with the current state of a company in terms of
energy performance and suggests practice measure to improve it.
232
The correlations focused on several key parameters that characterize companies of
233
agrifood sector, including: the quantity of raw material processed annually, the annual
234
electricity consumption, and the volume of cold rooms.
235
The energy performance algorithm is based in the research developed by Nunes et al.
236
[27; 30-32], who performed an intensive collection of field data on energy-related
237
characteristics of a large sample of agrifood industries.
238
The data was statistically organized, analysed, and interpreted in order to develop
239
analytical correlations that represent the Portuguese average energy performance of
240
meat, dairy and horticultural industries [27]. The statistical program SPSS version 18
241
was used to assess the data dispersion and to calculate the correlations. The
242
determination coefficient, R2, was used to set to correlation type. Based on the results, a
243
linear regression equation was used to obtain the relationships between the relevant
244
parameters. The regression outliers were removed from the sample using the
245
Chauvenet's criterion [33].
246
Figure 1 shows an example of the correlations developed for the sausages industry.
247
Figure 1 a) shows the relationship between raw material processed annually and
248
electricity consumption per year. Figure 1 b) shows the relationship between cold room
249
volume and electricity consumption per year.
250
251
Insert Figure 1 here
252
253
The correlations developed are shown in Table 1. These equations express the
254
relationships between the following relevant parameters: Electrical energy consumption
255
(annual), E [MhW]; Raw material (annual), RM [ton]; Compressors nominal power, CP
256
[kW] and Cold room volume, V [m3]. Note that the parameter of electricity consumption
257
accounts not only for the energy consumed for cooling or processing food products, but
258
also other sinks as lighting, heating, office, etc.
259
The electrical energy consumption per ton of raw material, SEC [kWh ton-1], on an
260
annual basis, is one of the power index among several indicators, which provides
261
information that characterizes the performance of agrifood companies in relation to their
262
electricity consumption. It provides the relationship between the electricity
263
consumption, E, and the processed raw material, RM. This index allows to perform the
264
comparison of the specific electrical energy consumption of similar processes on
265
different agrifood companies. The index evaluation allows to evaluate the performance
266
of agrifood companies and it cooling systems. The differences in this index value can be
267
used to find the causes of energy inefficiencies in agrifood companies by comparison of
268
equipment, infrastructure and processes. Table 1 also shows the average SEC values
269
from the sample for the different subcategories of agrifood companies.
270
271
Insert Table 1 here
272
273
As the correlations used in the algorithm are the core of the predictive tool and were
274
obtained through real values collected in companies, the predictions validity are limited
275
to a variation domain of each of characteristic parameter. The validity of the predictions
276
are constrained by the values given in Table 2.
277
278
279
Insert Table 2 here
280
3.3
Program flow
281
The predictive tool presented in this study considers improvements of the tool
282
developed by Santos et al. [34-35] and Neves et al. [36]. Initially, a window containing
283
a brief description of the predictive tool is displayed, informing the user about the
284
necessary input data (inputs) to perform the simulation in the following steps (see
285
Figure 2).
286
Insert Figure 2 here
287
288
289
As shown in Figure 3, in the next menu the user can select in which industry sector
290
fits the company to be analysed (see Figure 4-a)), including meat, fish, horticultural
291
(fruit and vegetables) and dairy products.
292
Insert Figure 3 here
293
294
295
There are multiple subcategories within each sector. In the specific case of the meat
296
sector industry, there are (1) Slaughterhouse, (2) Sausages industry and (3) Hams
297
industry as shown in Figure 4-b). Other sectors have also specific subcategories.
298
Pressing the button of any of the industry subcategories opens a window requesting
299
the inputs (Figure 4-c)). Here, the user must input the values of the characterization
300
parameters: amount of raw material processed annually [ton], annual electricity
301
consumption [MWh] and the total volume of cold rooms [m3]. Notice that it is always
302
possible to return to the previous menu or close the window using the respective buttons
303
(Back or Exit) for navigation.
304
305
Insert Figure 4 here
306
307
After entering the data according to the SI unit system and considering that the values
308
are within the valid variation domain of each parameter, when the button "Continue" is
309
pressed, the tool starts the computation based on the correlations and displays in a new
310
window the general results that summarize the current state of energy performance of a
311
company. In this window (see Figure 5), plots that relate the characterization parameters
312
of a company are displayed. All plots have a green shaded background that represents
313
the confidence level of 95% of the statistical correlations. In addition, it is also shown in
314
each plot the percent deviation of the parameter under consideration: (Company point –
315
red dot) and Portuguese national average (black dot). If the user wants to analyse a
316
particular plot, pressing the button above the plot will open a new and enlarged window.
317
Insert Figure 5 here
318
319
320
4 Case study
321
4.1
Companies presentation
322
To validate the predictive tool, three companies were selected, hereinafter referred as
323
Company A, Company B and Company C for reasons of business confidentiality, each
324
one for a subcategory of the meat industry: a slaughterhouse, a hams industry and a
325
sausages industry respectively. These companies have not been part of the statistical
326
data used in the development of correlations.
327
The company A is a slaughterhouse with 12 years of activity with 13 workers.
328
According to Portuguese legislation is classified as a micro-enterprise and is dedicated
329
to the meat production, processing annually 1473 tons of raw materials, particularly
330
cattle and pork meat. In its facilities there are nine cold rooms with a total volume of
331
638 m3 and its total covered area is 1117 m2. Some cold rooms are built in sandwich
332
panels and others in masonry insulated by polyurethane. The lighting of the cold rooms
333
is of fluorescent type and the refrigeration tubing are insulated by neoprene as shown in
334
Figure 6 a). The refrigeration compressors have a total rated nominal power of 43 kW
335
distributed evenly by four compressors Bitzer (direct central-circuit). The company
336
annual energy consumption is around 209 MWh. It has capacitors installed for the
337
correction of power factor for an actual value of 0.97. This company also has air-
338
conditioned corridors to avoid losses when accessing the cold rooms as shown in Figure
339
6 b). Additionally, it has a heat recovery unit and a diesel boiler.
340
Insert Figure 6 here
341
342
343
Company B manufactures hams. It has 24 years of experience, employing 15
344
technical workers. It has a covered area of 3000 m2. According to the Portuguese
345
legislation it is classified as a small-enterprise, dedicated to the production of meat,
346
annually processing about 2029.5 tons of feedstock (exclusively pork meat). In its
347
facilities there are 18 cold rooms (freezing, cooling and drying) with a total volume of
348
6668 m3. Eight cold rooms are made of masonry with cork insulation and the remaining
349
ten rooms are built in sandwich panel with polyurethane insulation (see Figure 7). The
350
lighting in the masonry made rooms’ uses incandescent lamps, and the others uses
351
fluorescent lamps. The corridors have no air-conditioning system although the facilities
352
include a heat recovery unit. The compressors of Company B have a total rated power
353
of 221 kW distributed by 18 compressors from the following brands: Bitzer, Copeland
354
and Dorin (individual cooling circuits). The annual global energy consumption is
355
1662.1 MWh and the company has a power factor of 0.91 corrected by capacitors. The
356
company also has a diesel boiler.
357
Insert Figure 7 here
358
359
360
Company C has 15 years of activity, employing only 4 workers. According to the
361
Portuguese legislation is classified as a micro-enterprise, devoted to the production of
362
dry-cured sausages. It processes annually 44.8 tons of raw material, mainly pork meat.
363
The company has three cold rooms with a total volume of 142 m3. However its covered
364
area is 600 m2 since the facilities include rooms for manufacturing the sausages and a
365
smokehouse for drying the products. The cold rooms are made with polyurethane
366
sandwich panels with fluorescent lighting (see Figure 8). The compressors have a total
367
rated power of 10.5 kW distributed by four Bitzer compressors (individual cooling
368
circuits) and the annual global energy consumption is 21.2 MWh. The power factor of
369
the company is 0.86. The facility has not a capacitors system for power factor
370
correction. This particular company does not have a stove either a boiler.
371
Insert Figure 8 here
372
373
374
The details of each industry are gathered in Table 3.
375
Insert Table 3 here
376
377
378
5 Analysis and discussion of results
379
The tool allows predicting the energy performance of a particular company in relation
380
to the Portuguese national average. Depending on the results, the tool also provides
381
practice measures to improve the energy efficiency. Thus, it can be used on energy-
382
decision making process in order to improve the energy efficiency if required and/or
383
desired.
384
This work shows the application of the predictive tool in case studies of industries in
385
the meat sector, particularly, in a slaughterhouse, a hams industry and a sausages
386
industry. The values of the parameters shown in Table 2 (company activity section) for
387
each case study were entered in the predictive tool.
388
Figures 9 to 11 show the overall results of the energy performance of each case study.
389
Based on the predictions, all companies have values of specific electrical energy
390
consumption below average, characterizing them as competitive businesses. In the case
391
of Company A, as shown in Figure 9, note that all plots show a small increment
392
comparing to the national average, such as the energy consumption vs. raw material,
393
where the value is 23% above average (see Figure 9 a). From this plot it appears that for
394
the annual energy consumption of the company, 209 MWh, it would be expected to
395
process more raw material (about 2000 tons). The plot relating the volume of cold
396
rooms vs. raw material predict that cold rooms are capable of storing more 23% of the
397
processed meat annually. An important feature is related to prediction values within the
398
confidence interval of 95% (shaded green). Finally, the plot of specific energy
399
consumption shows that this company has a SEC value 5% below the national average.
400
Therefore it is considered more efficient, having competitive advantage over others with
401
higher energy consumption, and thus being able to practice more competitive prices. If
402
the company processes more raw material, the value of the nominal power of
403
compressors in relation to the amount of raw material processed and volume of cold
404
rooms would be adjusted according to prediction correlation values. The compressors
405
power is 22% higher than expected, which allows concluding that this particular
406
industry can increase the volume of cold rooms or load (raw material) using the
407
installed power. However, based on the audit results there were identified certain issues
408
that may improve the efficiency of this industry, such as the replacement of fluorescent
409
lamps by LED lamps. Furthermore and according to the data collected there is in
410
average one person per hour inside the cold rooms. Despite the air conditioned systems
411
of hallways, this condition increase the thermal load of the cold rooms that lead to an
412
increased energy consumption due to increased working time of compressors to
413
compensate this thermal load. Another negative indicator was the poor condition of the
414
boiler’s insulation. On the other hand the power factor of the company assumes a value
415
close to ideal, being an asset to the energy performance of this industry.
416
Insert Figure 9 here
417
418
419
The company B is the largest from the case studies having a higher number of
420
technical staff than others. At a first glance, it is shown in Figure 10 that most of the
421
prediction values displayed in the plots for this industry are within the confidence
422
interval.
423
424
Insert Figure 10 here
425
426
In the first plot (see Figure 10 a), where energy consumption is related with the raw
427
material processed, it is predicted an energy consumption 31% lower than the average,
428
which allows pointing out that this is a competitive industry. Additionally, the
429
compressors nominal power for the same amount of raw material is 38% lower than the
430
national average. This condition arises from the fact that Company B process more raw
431
material than the national average for the same volume of cold rooms as shown in
432
Figure 10 c). Thus, the specific values related to the raw material processed are below
433
the national average resulting in a lower SEC. Regarding the energy consumption and
434
nominal power of compressors related to the volume of cold rooms (see Figure 10 e)
435
and Figure 10 f)), the values are 7% above and 12% below average, respectively. It
436
should be noted that these latter values lie within the confidence interval, so the
437
uncertainty of the predictions is reduced by the overall analysis of the correlations. The
438
SEC of company B is 32% below the national average being therefore the more efficient
439
company analysed in this case study. The cold storage of raw material is considered as a
440
great advantage in this kind of industry. During the audit, some inefficiencies were
441
detected, namely the lack of insulation in the boiler, the absence of air-conditioning in
442
the hallways and the type of illumination used. Regarding the power factor, the
443
capacitors system ensures that this value remains within the normal parameters.
444
The sausage industry examined within this case study is company C, the smallest one.
445
It is a micro company and the values of the input parameters for predicting the energy
446
performance are in the limit of validity domain of the results.
447
448
Insert Figure 11 here
449
450
Figure 11 a) shown an energy consumption 64% lower than the national average for
451
the amount of the raw material processed. In the remaining plots, the conclusions are
452
similar and all predicted values are below average. The compressors power and the
453
volume of cold rooms are lower than expected by 39% and 23% respectively for the raw
454
material processed when compared to the national average. Figure 11 e) and
455
Figure 11 f) compares the energy consumption and the nominal power of compressors
456
with the volume of cold rooms. In these plots is noted that the predicted values are
457
below the average 86% and 10% respectively. The prediction in Figure 11 e) (86%) is
458
outside the confidence interval due to an amount of raw material processed 23% lower
459
in relation to the volume of the cold room. This condition is verified by the low energy
460
consumption. Considering again Figure 11 a), it is clear that if the company handles
461
more quantity of raw materials, the energy consumption will increase. Finally, by
462
evaluating the plot comparing the SEC with the national average is predicted a value
463
28% lower than the estimated average. It is important to point out that this company
464
uses electricity for heating water. This condition contributes to the low power factor
465
(0.86) which increases the cost of electricity due to the increased reactive power. This
466
problem can be easily solved by installing a small capacitor system. Notice that
467
although the results obtained in the simulations are valid, the value of the amount of raw
468
materials is slightly below the valid domain what can compromise the prediction results,
469
hence the variation between the values of different plots (-86% maximum in Figure 11
470
e) and - 10% minimum in Figure 11 f).
471
472
The predictions results as well as the conclusions obtained from the audits are
473
indicative but allow managers of agrifood companies to be aware about the positioning
474
of their company in relation to the national average in terms of energy performance.
475
However, any change in the company or in the processes require a detailed in-situ study
476
about the energy consumption of different devices that are part of manufacturing and
477
cooling processes, taking into account the various inefficiencies that may exist.
478
Nevertheless, the results obtained by the tool for the prediction of energy performance
479
of meat industries are in line with the failures or inefficiencies detected in the fieldwork.
480
The predictive tool here presented can be adapted to another country or sector, requiring
481
only the modification of the correlations as well as the limit range of application and the
482
confidence intervals.
483
484
6 Conclusions
485
This paper presents a predictive tool of the energy performance of agrifood industry.
486
The main features and workflow of the tool are presented. Three case studies of its
487
application are provided. The selected industries belong each one to a different meat
488
subsector, respectively, slaughterhouse, hams industry and sausages industry. The case
489
studies validate and demonstrate the applicability of the predictive tool. Some
490
conclusions obtained from the graphical analysis of predicted results about the
491
positioning of each company performance in relation to the Portuguese national average
492
are presented. This predictive tool allows performing an evaluation of the energy
493
performance of companies in the agrifood sector comparing the most relevant
494
parameters such as raw materials processed annually, annual energy consumption and
495
volume of cold rooms with the national average. The predictions allows the user to
496
conclude which are the possible weaknesses or strengths of its company. The plot of
497
specific energy consumption is very conclusive since it relates the energy consumption
498
per ton of raw material. It can be seen as a competitiveness indicator. However it should
499
be mentioned that the developed tool only helps managers of agrifood industries in the
500
analysis of the energy performance. It is necessary that the user has sensitivity to
501
identify possible problems of technical origin on the company facilities. Thus, the
502
analysis does not discards the need for a more detailed study to determine particular
503
conditions that can be improved. This conclusion arises from the comparison of results
504
obtained from the simulations and the experimental results from audits.
505
The current state of the predictive tool allows the user to manually enter the data of
506
annual energy consumption, raw material processed annually and volume of cold
507
rooms. With this performance assessment, the user can decide how to improve the
508
energy performance of his company. The practical application of this tool demonstrates
509
its usefulness in helping decision-making in the implementation practice measures for
510
the improvement of energy efficiency. This tool includes also correlations for other
511
sectors of the agrifood sector, such as fish, milk, vine and wine and distribution.
512
Moreover, the predictive tool delivers automatic suggestions for the improvement of
513
energy performance depending on the prediction results. This tool can be used in any
514
country or sector, requiring only the modification of the correlations as well as the limit
515
range of application and the confidence intervals.
516
517
Acknowledgment
518
This work is part of the anchor project "InovEnergy – Eficiência Energética no Sector
519
Agroindustrial" inserted in the InovCluster plan of action. The study was funded by the
520
National Strategic Reference Framework (QREN 2007-2013), POFC (Operational
521
Programme for Competitiveness Factors), 01/SIAC/2011, Ref: 18642.
522
523
References
524
[1] Pachai, A.C.: "From Crandle to table – cooling and freezing of food". in 2nd IIR
525
International Conference on Sustainability and the Cold Chain (ICCC 2013), Paris,
526
France, 2-4 April, 2013.
527
[2] Laguerre, O., Hoang, H. M., Flick, D.: “Experimental investigation and modelling in
528
the food cold chain: Thermal and quality evolution”, Trends in Food Science &
529
Technology 29(2), pp. 87-97, 2013.
530
[3] IEA: 2012 Key World Energy Statistics, International Energy Agency (IEA), 2012.
531
[4] Gogou, E., Katsaros, G., Derens, E., Li, L., Alvarez, G., Taoukis, P.: "Development
532
and applications of the European Cold Chain Database as a tool for cold chain
533
management", in 2nd IIR International Conference on Sustainability and the Cold Chain
534
(ICCC 2013), Paris, France, 2-4 April, 2013.
535
[5] EC: FRISBEE project - Food Refrigeration Innovations for Safety, consumers
536
Benefit, Environmental impact and Energy optimisation along the cold chain in Europe.
537
European Commission (EC) - Research & Innovation.
538
http://ec.europa.eu/research/bioeconomy/food/projects/food_processing/frisbee_en.htm
539
(accessed in 18/02/14)
540
[6] Stahl, V., Alvarez, G., Derens, E., Hoang, H. M., Lintz, A., Hezard, B., El Jabri, M.,
541
Lepage, J. F. Ndoye, F. T., Thuault, D., Gwanpua, S. G., Verboven, P., Geeraerd, A.:
542
"An essential part of the frisbee software tool: Identification and validation of models
543
quantifying the impact of the cold chain on the RTE pork products", in 2nd IIR
544
International Conference on Sustainability and the Cold Chain (ICCC 2013), Paris,
545
France, 2-4 April, 2013.
546
[7] Foster, A., Zilio, C., Corradi, M., Reinholdt, L., Evans, J.: "Freely available cold
547
store energy models", in 2nd IIR International Conference on Sustainability and the
548
Cold Chain (ICCC 2013), Paris, France, 2-4 April, 2013.
549
[8] EC: ICE-E Improving cold storage equipment in europe, European Commission
550
(EC), Energy. http://eaci-projects.eu/iee/page/Page.jsp?op=project_detail&prid=2365
551
(accessed in 18/02/14)
552
[9] Evans, J. A., Hammond, E. C., Gigiel, A. J., Foster, A. M., Reinholdt, L., Fikiin, K.,
553
Zilio, C.: “Assessment of methods to reduce energy consumption of food cold stores”,
554
Applied Thermal Engineering 62(2), pp. 697–705, 2014.
555
[10] Evans, J.A., Foster, A. M., Huet, J.-M., Reinholdt, L., Fikkin, K., Zilio, C.,
556
Houska, M., Landfeld, A., Bond, C., Scheurs, M., T.W.M. van Sambeeck: “Specific
557
energy consumption values for various refrigerated cold stores”, Energy and Buildings
558
74, pp. 141-151, 2014.
559
[11] Eden, M., Colmer, C.: “Improved Cold Chain Management: efficiency and food
560
safety through international project results”, Food Safety and Tecnhology 24(1), pp. 30-
561
32, 2010.
562
[12] EC: CHILL-ON – Enhancing safety at every step on the food supply chain,
563
European
564
http://ec.europa.eu/research/infocentre/article_en.cfm?id=/research/star/index_en.cfm?p
565
=ss-chill-on&item=Infocentre&artid=25333 (accessed in 18/02/14)
Commission
(EC),
Research
&
Innovation.
566
[13] Colmer, C., Kuck, M., Lohmann, M., Bunke, M.: "Novel Technologies to improve
567
safety and transparency of the chilled food supply chain", Innovation in the nordic
568
marine sector, May, 2009.
569
[14] Van der Sluis, S.M.: “Wind energy storage in cold stores: Night Wind”, in Energy
570
Efficiency & Renewable Energy Sources (EE & RES), 4th International Congress for
571
South-East Europe, Sofia, Bulgaria, 7-9 April, 2008.
572
[15] Saint Trofee: Refrigeration Tools, http://www.nightwind.eu/night-wind.html
573
(accessed in 18/02/14)
574
[16] Ge, Y.T., Tassou, S.A.: “Control optimizations for heat recovery from CO2
575
refrigeration systems in supermarket”, Energy Conversion and Management 78, pp.
576
245–252, 2014.
577
[17] Romero, J.A., Navarro-Esbrí, J., Belman-Flores, J.M.: "A simplified black-box
578
model oriented to chilled water temperature control in a variable speed vapour
579
compression system", Applied Thermal Engineering 31(2–3), pp. 329-335, 2011.
580
[18] González-Ramírez, J.E., Leducq, D., Arellano, M., Alvarez, G.: “Energy
581
consumption optimization of a continuous ice cream process“, Energy Conversion and
582
Management 70, pp. 230–238, 2013.
583
[19] Subiantoro, A., Ooi, K.T.: "Economic analysis of the application of expanders in
584
medium scale air-conditioners with conventional refrigerants, R1234yf and CO2",
585
International Journal of Refrigeration 36(5), pp. 1472-1482, 2013.
586
[20] Antonijevic, D.: "An engineering procedure for air side performance evaluation of
587
flat tube heat exchangers with louvered fins", Heat and Mass Transfer 49(1), pp. 117-
588
127, 2013.
589
[21] Khan, M.S., Lee, S., Rangaiah, G.P., Lee, M.: Knowledge based decision making
590
method for the selection of mixed refrigerant systems for energy efficient LNG
591
processes", Applied Energy 111, pp. 1018-1031, 2013.
592
[22] AlQdah, K.: “Potential opportunities for energy savings in a Jordanian poultry
593
company”, Energy Conversion and Management 51, pp. 1651–1655, 2010.
594
[23] Evans, J.A., Hammond, E.C., Gigiel, A.J., Reinholdt, L., Fikiin, K., Zilio, C.:
595
"Improving the energy performance of cold stores", in 2nd IIR International Conference
596
on Sustainability and the Cold Chain (ICCC 2013), Paris, France, 2-4 April, 2013.
597
[24] IAPMEI: PME em números, http://www1.ionline.pt/conteudo/37754-populacao-
598
activa-portuguesa-e-das-que-tem-menos-habilitacoes-literarias- (in Portuguese)
599
[25] Santiago, P., Donaldson, G., Looney A., Nusche, D.: “OECD Reviews of
600
Evaluation and Assessment in Education: Portugal 2012”, Organization for Economic
601
Cooperation Development (OECD) Publishing, 2012.
602
[26] INE: Statistical Yearbook of Portugal, Instituto Nacional de Estatística (INE),
603
Lisbon, Portugal, 2009.
604
[27] Nunes, J.: "Energetic efficiency evaluation in refrigeration systems of agrifood
605
industries in the Beira Interior region", PhD Thesis, University of Beira Interior,
606
Covilhã, Portugal, 2014.
607
[28] Matworks: MATLAB Primer, The Mathworks, Inc., USA, 2013.
608
[29] Matworks: MATLAB Creating Graphical User Interfaces, The Mathworks, Inc.,
609
USA, 2013.
610
[30] Nunes, J., Silva, P.D., Andrade, L.P.: "Energetic efficiency evaluation in
611
refrigeration systems of meat industries", in 23rd International Congress of
612
Refrigeration (ICR 2011), Prague, Czech Republic, August 21 - 26, 2011.
613
[31] Nunes, J., Silva, P.D., Andrade, L.P., Gaspar, P.D.: "Characterization of specific
614
energy consumption of electricity in the Portuguese sausages industry", Energy and
615
Sustainability - WIT Transactions on Ecology and the Environment, vol. 186, 2014.
616
[32] Nunes, J., Silva, P.D., Andrade, L.P., Gaspar, P.D., Domingues, L.C.: "Energetic
617
evaluation of refrigeration systems of horticultural industries in Portugal", in 3rd IIR
618
International Conference on Sustainability and Cold Chain (ICCC 2014), London,
619
United Kingdom, June 23-25, 2014.
620
[33] ASHRAE: Engineering Analysis of Experimental Data - Ashrae Guideline 2,
621
American Society of Heating, Refrigerating and Air-Conditioning Engineers
622
(ASHRAE), Atlanta, 1986.
623
[34] Santos, R., Nunes, J., Silva, P.D., Gaspar, P.D., Andrade, L.P.: "Ferramenta
624
computacional de análise e simulação do desempenho de unidades de conservação de
625
carne através de frio industrial", in VI Congreso Ibérico y IV Congreso Iberoamericano
626
de Ciencias y Técnicas del Frío (CYTEF-2012), Madrid, Spain, February, 2012.
627
[35] Santos, R., Nunes, J., Silva, P.D., Gaspar, P.D., Andrade, L.P.: "Predictive tool for
628
the analysis and simulation of cold room performance in perishable products industry",
629
in 2nd IIR International Conference on Sustainability and the Cold Chain (ICCC 2013),
630
International Institute of Refrigeration (IIR), Paris, France, April 2-4, 2013.
631
[36] Neves, D., Gaspar, P.D., Silva, P.D., Andrade, L.P., Nunes, J.: “Predictive tool for
632
the energy efficiency assessment of cheese industries - Case study of inner region of
633
Portugal”, in V Congreso Iberoamericano de Ciencias y Técnicas del Frío (CYTEF
634
2014), Tarragona, Spain, June 18 - 20 2014.
635
Tables caption:
636
637
Table 1 – Statistical correlation for the energy performance of agrifood companies of the meat
638
sector.
639
Table 2 - Application domain of each parameter of the predictive tool.
640
Table 3 – Companies characterization and activity.
641
642
643
Figures caption:
644
645
646
647
648
Figure 1 – Example of the correlations developed for the sausages industry.
a) Relationship between raw material processed annually and electricity
consumption per year.
b) Relationship between cold room volume and electricity consumption per year.
649
Figure 2 – Cool-OP: Opening window.
650
Figure 3 – Cool-OP menu tree.
651
Figure 4 – Cool-OP: Windows to input the company parameters.
652
a) Menu window.
653
b) Menu window of meat sector.
654
c) Input window of the parameters of the industry (ex: slaughterhouse).
655
Figure 5 – Cool-OP: Window with the overall results (blank).
656
Figure 6 – Company A facilities.
657
a) example of evaporators’ type and location and lighting type of a cold room.
658
b) example of an air-conditioned corridor.
659
Figure 7 – Company B facilities: example of a cold room storing hams.
660
Figure 8 – Company C facilities: example of a cold room storing dry-cured sausages.
661
Figure 9 – Company A: Predictions of energy performance.
662
a) Electrical energy consumption vs Raw material.
663
b) Compressors nominal power vs. Raw material.
664
c) Cold room volume vs. Raw material.
665
d) Specific energy consumption.
666
e) Electrical energy Consumption vs. Cold room volume.
667
f) Compressors nominal power vs. Cold room volume.
668
Figure 10 – Company B: Predictions of energy performance.
669
a) Electrical energy consumption vs Raw material.
670
b) Compressors nominal power vs. Raw material.
671
c) Cold room volume vs. Raw material.
672
d) Specific energy consumption.
673
e) Electrical energy Consumption vs. Cold room volume.
674
f) Compressors nominal power vs. Cold room volume.
675
Figure 11 – Company C: Predictions of energy performance.
676
a) Electrical energy consumption vs Raw material.
677
b) Compressors nominal power vs. Raw material.
678
c) Cold room volume vs. Raw material.
679
d) Specific energy consumption.
680
e) Electrical energy Consumption vs. Cold room volume.
681
f) Compressors nominal power vs. Cold room volume.
682
683
Table 1 – Statistical correlation for the energy performance of agrifood companies of the meat sector.
Industry
Relationships
Electricity consumption, E [MhW] vs. Raw material, RM [ton]
Compressors nominal power, CP [kW] vs. Raw material, RM
[ton]
Cold room volume, V [m3] vs. Raw material, RM [ton]
Slaughterhouses
Sausages
Hams
E = 0.2041 RM - 138.97
E = 0.2376 MP + 24.18
E = 0.9945 RM + 159.93
(R2 = 0.57)
(R2 = 0.92)
(R2 = 0.86)
CP = 0.0520 RM - 51.46
CP = 0.0702 RM + 11.49
CP = 0.1082 RM + 86.20
(R2 = 0.71)
(R2 = 0.68)
(R2 = 0.73)
V = 0.5151 RM - 265.58
V = 4.3795 RM - 87.42
V = 3.8629 RM + 785.44
(R2 = 0.90)
(R2 = 0.82)
(R2 = 0.89)
E = 0.4557 V - 112.71
E = 0.3243 V - 6.84
E = 0.2667 V – 228.61
(R2 = 0.84)
(R2 = 0.96)
(R2 = 0.96)
3
Electricity Consumption, E [MWh] vs. Cold room volume, V [m ]
684
685
CP = 0.1102 V - 36.72
CP = 0.0572 V + 3.41
CP = 0.0323 V + 32.05
Compressors nominal power, CP [kW] vs. Cold room volume, V
[m3]
(R2 = 0.94)
(R2 = 0.93)
(R2 = 0.90)
Average specific energy consumption, SEC [kWh ton-1]
148.5
660.2
1208.0
Table 2 – Application domain of each parameter of the predictive tool.
686
687
688
Raw Material
Volume of cold rooms
Compressors nominal power
[ton]
[m3]
[kW]
Slaughterhouses
1000-5000
500-3000
25-300
Sausages
50-1100
10-1000
10-100
Hams
50-2500
1500-11000
80-350
Industry
689
Table 3 – Companies characterization and activity.
690
Company activity
Company
Company A
Company B
Company C
Slaughterhouses
Hams industry
Sausages industry
Quantity of raw material [ton]
1473.0
2029.5
44.8
Electricity consumption [MWh]
209.0
1662.1
21.2
Total volume of cold rooms [m3]
638
6668
142
Compressors nominal power [kW]
43
221
10.5
Company age
12
24
15
Number of workers
13
15
4
Micro company
Small company
Micro company
9
18
3
1117
3000
600
Fluorescent
Fluorescent/Incandescent
Fluorescent
Neoprene
Neoprene
Polyurethane
Polyurethane
Polyurethane/cork
Polyurethane
Capacitors battery
Yes
Yes
No
Power factor
0.97
0.91
0.86
Boiler
Yes
Yes
No
Air-conditioned corridors
Yes
No
No
Meat subsector
Company characterization
Company size
691
Number of cold rooms
Covered area [m2]
Lighting
Tube insulation
Chambers insulation
300
Electricity consumption [MWh]
Electricity consumption [MWh]
300
250
E = 0.2376 RM + 24.18
R² = 0,92
200
150
100
50
0
0
200
400
600
800
Raw material [ton]
1000
1200
250
E = 0.3243 V - 6,84
R² = 0.96
200
150
100
50
0
0
200
400
600
Volume [m3]
b) Relationship between cold room volume and electricity consumption
consumption per year.
per year.
693
694
696
1000
a) Relationship between raw material processed annually and electricity
692
695
800
Figure 1 – Example of the correlations developed for the sausages industry.
697
698
699
700
701
702
Figure 2 – Cool-OP: Opening window.
Figure 3 – Cool-OP menu tree.
c) Input window of the parameters
b) Menu window of
a) Menu window.
of the industry (ex:
meat sector.
slaughterhouse).
Figure 4 – Cool-OP: Windows to input the company parameters.
Figure 5 – Cool-OP: Window with the overall results (blank).
a) example of evaporators’ type and location and lighting type of a cold
room.
Figure 6 – Company A facilities:
b) example of an air-conditioned corridor.
Figure 7 – Company B facilities: example of a cold room storing hams.
Figure 8 – Company C facilities: example of a cold room storing dry-cured sausages.
a) Electrical energy consumption vs. Raw material.
d) Specific energy consumption.
b) Compressors nominal power vs. Raw material.
c) Cold room volume vs. Raw material.
e) Electrical energy Consumption vs. Cold room
volume.
Figure 9 – Company A: Predictions of energy performance.
f) Compressors nominal power vs. Cold room
volume.
a) Electrical energy consumption vs. Raw material.
d) Specific energy consumption.
b) Compressors nominal power vs. Raw material.
c) Cold room volume vs. Raw material.
e) Electrical energy Consumption vs. Cold room
f) Compressors nominal power vs. Cold room
volume.
volume.
Figure 10 – Company B: Predictions of energy performance.
a) Electrical energy consumption vs. Raw material.
d) Specific energy consumption.
b) Compressors nominal power vs. Raw material.
c) Cold room volume vs. Raw material.
e) Electrical energy Consumption vs. Cold room
f) Compressors nominal power vs. Cold room
volume.
volume.
Figure 11 – Company C: Predictions of energy performance.
*Detailed Response to Reviewer Comments
Response to reviewers
Journal: Energy Conversion and Management
Ref: ECM-D-14-01362
Title: Predictive tool of energy performance in agrifood industries: The portuguese case study
Reviewer #1:
1. It may be more appropriate if the title of the paper is changed as follows: Predictive tool of
energy performance in cold storage: The portuguese case study
[Authors answer] The title of the paper was changed according the reviewer suggestion, but
relating it to agrifood industries.
2. The aim of the study could be clearly written to the end of the introduction.
[Authors answer] A paragraph was inserted at the end of the Introduction section to clearly
identify the aim of the study according to reviewer suggestion.
Reviewer #2:
Authors present a predictive tool for the assessment of energy efficiency in the agrifood industry, I
believe the work is well written and has merit, however I would suggest to take into consideration
the following aspects.
1. I think the major contribution of the paper is the predictive tool (the algorithm) not the visual
interface, thus I would like to see more detail on such algorithm, the related correlations and the
statistic tools used in the study. This way, and as authors stated, the tool can be used by any
interested user in any given location.
[Authors answer] A new section "Materials and Methods" has been included in the manuscript.
This section describes in detail the sample characterization, the data collection procedure and the
equipment and measurement techniques used during the energy audits. Additionally, the
procedure how the data was statistically organized, analysed, and interpreted using the SPSS
software is described. This software was used to extract the analytical correlations that represent
the Portuguese average energy performance of meat, dairy and horticultural industries. In this
context, new example figures showing the correlations developed for the sausages industry are
included in the manuscript. A new table is included in the manuscript presenting the equations
that express the relationships between the relevant analysed parameters: Electrical energy
consumption (annual); Raw material (annual); Compressors nominal power and Cold room
volume.
2. Related to the previous comment, I believe it is not necessary to describe what a GUIDE is, or
the program flow. The user interface appears to be user friendly and straightforward to use,
instead include a detailed description of the insights of the predictive tool and how to determine
the required correlations.
[Authors answer] The detailed description of the insights of the predictive tool and how to
determine the required correlations were included in the manuscript as already described.
3. Tables 2 and 3 can be merged, as they contain information related to the industries in the cases
study.
[Authors answer] Tables 2 and 3 were merged according to reviewer suggestion.
4. Can you further explain why in industry B the electrical power consumption and B the nominal
power of the compressors are 31% and 38% lower than the national average? These values do
not lie within the confidence interval, are they statistically significant?
[Authors answer] The reason is due to the company B process more raw materials than the
national average for the same volume of cold rooms as shown in Figure 10 c).
Thus the specific values related to the raw material processed (Electrical energy consumption vs.
Raw material; Compressors nominal power vs. Raw material; and Cold room volume vs. Raw
material) are below the national average resulting in a lower SEC. The manuscript was improved
to accommodate this additional explanation.
In fact, the behaviour of this company regarding the electrical power consumption parameters and
the compressors nominal power are outside the confidence interval. Although, the uncertainty of
the predictions are reduced by the overall analysis of the correlations. The predictions of the
energy consumption and compressors nominal power related to the cold rooms volume lie within
the confidence interval strengthening the confidence in the predictions of energy consumption and
nominal power of compressors in relation to the raw material processed.
Reviewer #3:
The paper has been studied thoroughly and the comments are as follow:
The language level is good but, technically, the paper is not easy to understand and the way it is
presented, may not be acceptable for publication. Methodology of the research work is missing.
There is no equation that shows how you correlated various parameters. Some figures are trivial
and does not provide any information (technical or scientific). Graphs are reported and are not
technically discussed on scientific evidences.
Due to above mentioned reasons, I may not be in a position to recommend this paper for
publication.
[Authors answer] The manuscript was deeply reviewed taking into account the reviewer
comments. A new section "Materials and Methods" has been included in the manuscript. This
section describes in detail the methodology of the research work, including the sample
characterization, the data collection procedure and the equipment and measurement techniques
used during the energy audits. Additionally, the procedure how the data was statistically
organized, analysed, and interpreted using the SPSS software is described. This software was used
to extract the analytical correlations that represent the Portuguese average energy performance of
meat, dairy and horticultural industries. In this context, new example figures showing the
correlations developed for the sausages industry are included and technically discussed in the
manuscript. A new table is included in the manuscript presenting the equations that express the
relationships between the relevant analysed parameters: Electrical energy consumption (annual);
Raw material (annual); Compressors nominal power and Cold room volume.
*Complete Manuscript including All Figs - Marked version
1
Predictive tool of energy performance of cold storage in agrifood
2
industries: The portuguese case study
3
4
José Nunes1, Diogo Neves2, Pedro D. Gaspar2,*, Pedro D. Silva2, and Luís P. Andrade1
1
5
Politechnic Institute of Castelo Branco, Av. Pedro Álvares Cabral nº12 6000 Castelo
6
Branco, Portugal
7
[email protected], [email protected]
8
2
University of Beira Interior, Electromechanical Engineering Department, Rua Marquês
9
d’Ávila e Bolama 6200-001 Covilhã, Portugal
10
[email protected], [email protected] (* corresponding author), [email protected]
11
12
Abstract
13
Food processing and conservation represent decisive factors for the sustainability of
14
the planet given the significant growth of the world population in the last decades.
15
Therefore, the cooling process during the manufacture and/or storage of food products
16
has been subject of study and improvement in order to ensure the food supply with good
17
quality and safety. A predictive tool for assessment of the energy performance in
18
agrifood industries that use cold storage is developed in order to contribute to the
19
improvement of the energy efficiency of this industry. The predictive tool is based on a
20
set of characteristic correlated parameters: amount of raw material annually processed,
21
annual energy consumption and volume of cold rooms. Case studies of application of
22
the predictive tool consider industries in the meat sector, specifically slaughterhouses.
23
The results obtained help on the decision-making of practice measures for improvement
24
of the energy efficiency in this industry.
25
26
27
Keywords: Energy efficiency; sustainability; perishable products; cold storage;
predictive tool; cold industry.
28
29
1
Introduction
30
The sustainability and food safety in the agrifood sector has been a topic of research
31
and study. In the early twentieth century, the world population was about 1500 million
32
people, however just over a century the number of habitants increased to approximately
33
7000 million [1]. Thus, it is clear that the world demand for food is increasing which is
34
reflected on an enormous strain for the production and conservation sectors of the food
35
chain. So it becomes imperative to find short-term solutions for the agrifood industries
36
that in a macroscopic vision will help the sustainability of the planet. In this context,
37
cooling plays an important role, allowing to store food products in times of increased
38
production when the market has no capacity for product flow, or just make them
39
available when needed. The refrigeration process has the ability to preserve perishable
40
products ensuring their chemical, physical and nutritional properties but it is also
41
indispensable in the processing of perishable foods such as meat, fish, milk, fruit and
42
other. The demand of chilled and frozen food as increased substantially, which leads to
43
new requirements for the efficiency of refrigeration systems [2]. Energy represents a
44
factor of greatest importance not only for a country economy, but especially for the
45
well-being of its citizens. In 1971 there was a global electricity consumption of around
46
439 Mtone, and in 2010 this figure rose to 1.536 Mtone [3]. This increase is related to
47
the population growth, being the industry responsible for about 35.2% of energy
48
consumption worldwide in 2010. The energy consumed globally for the cooling process
49
accounts for 15% of energy consumption [1].
50
Thus, it is important to develop studies and tools that help the efficiency
51
improvement of these processes to ensure a better sustainability of this industrial sector.
52
There are some results provided by studies and projects in this area, more specifically in
53
the development of predictive tools for the analysis of several points related to the
54
cooling processes (whether energy consumption, environmental impact, among others).
55
The tool FRISBEE CCP (Cold Chain Predictor) [4-6], developed within the activities
56
of FRISBEE (Food Refrigeration Innovations for Safety, Consumers Benefit,
57
Environmental impact and Energy) project, performs simulations on specific conditions
58
defined by the user, constructs graphs representing the variation of temperature over
59
time and predicts the remaining shelf time of the product. These simulations are
60
performed based on the method of Monte Carlo [4] generating distributions of
61
time/temperature for every stage of the cold-chain and selected product. The results
62
represent realistic scenarios for the behaviour of food products, being possible to take
63
corrective actions in order to optimize the efficiency of the cold-chain ensuring product
64
quality and increasing its validity. This tool is intended to provide free information and
65
tools to managers, designers and operators of the refrigeration industry optimizing the
66
efficiency of their companies. In this same context, Foster et al. [7] describes the ICE-E
67
(Improving Cold Storage Equipment in Europe) project [8] devoted to the development
68
of tools with the same goals. This project contributed to reduce the energy consumption
69
and greenhouse gas emissions from the European food cold storage sector through the
70
application of more energy efficient equipment and taking into account the energy and
71
environmental standards of the EU [8-10].
72
Eden et al. [11] describes a predictive tool, QMRA (Quantitive Microbial Risk
73
Assessment) [12-13], which is combined with the principle of the Hazard Analysis and
74
Critical Control Points (HACCP) to enable the improvement of quality and safety of
75
food in a preventive approach. More specifically, this tool allows to estimate the risk of
76
pathogens growth based on the temperature and chemical and nutritional characteristics
77
of food. As result, the SLP (Shelf Life Predictor) [13] was developed with the ability to
78
predict the true and the remaining shelf life of foods. Apart from the quality and safety
79
of foods, this tool allows tracking (traceability) products in order to know the places
80
where they are in real time so that the consumer can make use of reliable information
81
about the origin of food. It also allows to overlook the manufacturing process from the
82
collection of raw materials to their processing, helping to ensure quality and safety.
83
Van der Sluis [14] describes a software that makes the temperature control of cold
84
stores taking into account the price of electricity and the daily consumption profile of
85
the company. The load management was accomplished considering that compressors
86
will work during time periods of low-cost electricity due to lower demand, in order to
87
accumulate energy in the form of cold to rationally use it during peak hours [15].
88
However, this procedure is executable only if the facilities have enough cold storage
89
capacity to store the energy demand during the day. Besides the abovementioned works,
90
other computational tools and models have been developed to enhance the operation of
91
refrigeration facilities [16] and system devices, such as compressors and their control,
92
regulation and command [17-18], expanders [19], among others. Additionally,
93
methodologies, strategies and procedures has been developed for optimal control of heat
94
recovery processes in refrigeration systems [16], for the performance evaluation of heat
95
exchangers [20] as well as knowledge based decision making methods for refrigerant
96
selection [21] and energy savings based in energy audits and its detailed analysis [22].
97
These works are only few examples of recent ongoing research and development
98
focusing the energy improvement of refrigeration systems, that are the main consumers
99
in the agrifood industry.
100
This paper presents a predictive tool developed for the assessment of the energy
101
performance in agrifood industries that use cold storage. The tool uses mathematical
102
correlations from the greatest number of data sets collected to date in Portugal.
103
Statistical correlations between raw material, energy consumption and volume of cold
104
rooms are used. To show the features and applicability of the tool, several case studies
105
for different subsectors of the meat industry are analysed. The tool results will be useful
106
in the decision-making process of practice measures for the improvement of energy
107
efficiency.
108
109
2
Material and Methods
110
2.1
Framework in the context of Portuguese industries
111
The abovementioned research covers the development of predictive tools aimed to
112
improve the energy efficiency of cold stores since refrigeration is responsible for about
113
60-70% of electricity consumption of these facilities [23]. Although, Portuguese
114
companies had not been included in the databases. In this context, a project was
115
developed in Portugal focussed in the identification of energy consumption profiles in
116
the agrifood industry. Additionally, it intends to promote and develop actions that
117
contribute to a real improvement of energy efficiency and competitiveness of this
118
sector. The work developed and presented in this paper is part of a project activity and
119
its main objective is the development and implementation of an analysis algorithm to be
120
validated with companies outside the sample of technical audits. As part of the activity,
121
it is aimed the development of a predictive tool to support strategic decisions in
122
companies allowing the estimation of their energy performance and pointing practice
123
measure that lead to an effective improvement of energy efficiency. Note that this
124
project is not aimed for the energy characterization of the cold stores in particular, but to
125
the energy characterization by agrifood category, i.e., meat, fish, dairy, horticultural
126
(fruit and vegetables), wine and vineyard and distribution, in order to obtain real data
127
that was used for the development of the correlations included in the model/algorithm.
128
The Portuguese industry is mainly constitute by small and medium enterprises. In
129
2011, it number was around 1.112.000, providing employment to about 78.5% of the
130
Portuguese population [24-25]. A large part of these industries are included in the
131
agrifood sector.
132
133
2.2
Sample
134
This study was conducted based on a sample of 87 agrifood companies: 33 from the
135
meat sector, 31 from the dairy sector and 23 from the horticultural (fruits and
136
vegetables) sector in the central region of Portugal. According to INE [26], this region
137
has the highest number of agrifood companies in Portugal. Additionally, this region has
138
the perfect weather conditions and soil and vegetation properties that allows the
139
development of agrifood practices for a wide range of products from different sectors.
140
Several regional products are manufactured in the region, among which stand out the
141
products in the meat sector (cured meats), dairy products (cheeses) and fruits.
142
In particular, the meat sector industries, where the predictive tool was applied and
143
validated, were subdivided into three subcategories: Slaughterhouse, Sausages industry
144
and Hams industry. The analysis was conducted in 4 Slaughterhouses plants, 20
145
Sausages industries and 9 Hams industries. 70% of the ham manufacturing companies
146
are located in this region of the country.
147
148
2.3
Data Collection
149
Initially, an extensive survey was developed to carry out a collection of information
150
about the agrifood industries. This survey was used to collect information regarding
151
physical characteristics of plants, activities and characteristics of production processes,
152
high energy-use equipment with special emphasis on cooling systems, cooling and
153
freezing chambers and energy consumption. Besides serving as a support document for
154
data recording, the document ensured a systematic and reliable data collection.
155
Moreover, it was also used as a technical guide to the plants inspection. The inquiry
156
document was used to record the following data:
157
i) Identification and location of agrifood plants;
158
ii) Characterization of infrastructure (location, age, type of materials, dimensions and
159
160
161
location of cooling and freezing chambers);
iii) Characterization of activities (type and quantities of raw materials and final
products);
162
iv) Identification and characterization of the production process;
163
v) Identification and quantification of energy types;
164
vi) Characterization of the electricity tariff and consumption (tariff type, electrical
165
power contracted, electricity consumption, power factor);
166
vii) Breakdown of electricity consumption by sector and equipment;
167
viii) Characterization of cooling and freezing chambers (number, type of insulation
168
169
170
material, preservation condition, raw materials/products load);
ix) Determination of internal and external environmental conditions, of facilities and
cooling chambers (air temperature and relative humidity);
171
x) Identification and characterization of refrigeration systems (age, location, type of
172
technology and technical characteristics, refrigerant type and nominal electrical powers
173
of compressors);
174
xi) Evaluation of the characteristics of compressed air systems;
175
xii) Characterization of thermal fluid generators;
176
xiii) Evaluation of energy efficiency improvements.
177
The collected information relates to activities undertaken during 2008.
178
The data collection process was carried out face-to-face, with a visit to the industries,
179
and where the building, the equipment, the technical operations and the manufacturing
180
processes were observed. Data collection was performed during the year 2009.
181
182
2.4
Equipment and measurement techniques
183
In order to measure the environmental conditions inside and outside the cooling
184
equipment, including air temperature a006Ed moisture content, a digital multi-function
185
measuring equipment (Testo 435-2) was used, along with its temperature and humidity
186
probes (±0.3°C accuracy). The previous equipment with a contact probe with type K
187
thermocouple (±0.5°C accuracy) was used to measure the walls/surfaces temperature.
188
Temperature data-loggers (Microlite and Lascar electronics with (±0.5°C accuracy)
189
were used to measure the air temperature inside the cooling and freezing chambers. A
190
thermographic camera Testo 880 (±2°C accuracy) was used to evaluate the preservation
191
condition of cooling chambers walls, thermal bridges, sealing condition of chambers
192
doors, and insulation of refrigerant suction pipes and of discharge air ducts of air
193
handling units. An ammeter clamp was used for measuring the electrical current and
194
voltage input to the compressor motor, in order to evaluate its electrical power
195
consumption. The electrical power data obtained experimentally was compared with the
196
estimated values available from official catalogues and dedicated software, for the same
197
operating conditions. Close results (deviations below 10%) were found between the two
198
methods in use. Although the energy intakes were quantified by analysis of the monthly
199
electricity bills, a profile analysis of electricity consumption was also performed using
200
an energy analyser (Elcontrol Energy Explorer).
201
202
203
The physical dimensions of the cooling and freezing chambers were calculated using
an infra-red distance meter Bosch-DLE 40 (± 1.5 mm accuracy).
204
2.5
Database
205
A computer database was built to record all information gathered from the plants.
206
This database was designed to allow data processing and analysis. It also made possible
207
to obtain various performance indicators from each agrifood company and
208
simultaneously to carry out a comparative analysis between different plants in each
209
sector. Additional details of the database can be found in [27].
210
211
3
Predictive tool features
212
3.1
Overview
213
The predictive tool, Cool-OP (Cooling Optimization Program), was developed in
214
MATLAB [28] making use of GUIDE (Graphical User Interface Design Environment)
215
[29]. This tool provides point-and-click control of software applications, reducing the
216
need to type commands in order to run an application. GUIDE provides tools for
217
graphically design the user interface for applications. The graphical user interface
218
typically contains controls named components such as menus, toolbars, radio and push
219
buttons, edit text, static text and check boxes, sliders, among others. Then, GUIDE
220
automatically generates the MATLAB code for the user interface, which was
221
programmed with the behaviour expected for the application. The computations to be
222
performed by the application were programmed in MATLAB language and the data is
223
display in plots.
224
The user interface of the predictive tool is extremely simple, intuitive, user-friendly
225
and easy to understand. The aim is to provide a tool accessible to all professionals in the
226
agrifood industry regardless of their qualifications. Finally, the code generated in
227
MATLAB was converted to an executable file that can run in any computer.
228
229
230
231
3.2
Algorithms
The predictive tool provides the user with the current state of a company in terms of
energy performance and suggests practice measure to improve it.
232
The correlations focused on several key parameters that characterize companies of
233
agrifood sector, including: the quantity of raw material processed annually, the annual
234
electricity consumption, and the volume of cold rooms.
235
The energy performance algorithm is based in the research developed by Nunes et al.
236
[27; 30-32], who performed an intensive collection of field data on energy-related
237
characteristics of a large sample of agrifood industries.
238
The data was statistically organized, analysed, and interpreted in order to develop
239
analytical correlations that represent the Portuguese average energy performance of
240
meat, dairy and horticultural industries [27]. The statistical program SPSS version 18
241
was used to assess the data dispersion and to calculate the correlations. The
242
determination coefficient, R2, was used to set to correlation type. Based on the results, a
243
linear regression equation was used to obtain the relationships between the relevant
244
parameters. The regression outliers were removed from the sample using the
245
Chauvenet's criterion [33].
246
Figure 1 shows an example of the correlations developed for the sausages industry.
247
Figure 1 a) shows the relationship between raw material processed annually and
248
electricity consumption per year. Figure 1 b) shows the relationship between cold room
249
volume and electricity consumption per year.
250
251
Insert Figure 1 here
252
253
The correlations developed are shown in Table 1. These equations express the
254
relationships between the following relevant parameters: Electrical energy consumption
255
(annual), E [MhW]; Raw material (annual), RM [ton]; Compressors nominal power, CP
256
[kW] and Cold room volume, V [m3]. Note that the parameter of electricity consumption
257
accounts not only for the energy consumed for cooling or processing food products, but
258
also other sinks as lighting, heating, office, etc.
259
The electrical energy consumption per ton of raw material, SEC [kWh ton-1], on an
260
annual basis, is one of the power index among several indicators, which provides
261
information that characterizes the performance of agrifood companies in relation to their
262
electricity consumption. It provides the relationship between the electricity
263
consumption, E, and the processed raw material, RM. This index allows to perform the
264
comparison of the specific electrical energy consumption of similar processes on
265
different agrifood companies. The index evaluation allows to evaluate the performance
266
of agrifood companies and it cooling systems. The differences in this index value can be
267
used to find the causes of energy inefficiencies in agrifood companies by comparison of
268
equipment, infrastructure and processes. Table 1 also shows the average SEC values
269
from the sample for the different subcategories of agrifood companies.
270
271
Insert Table 1 here
272
273
As the correlations used in the algorithm are the core of the predictive tool and were
274
obtained through real values collected in companies, the predictions validity are limited
275
to a variation domain of each of characteristic parameter. The validity of the predictions
276
are constrained by the values given in Table 2.
277
278
279
Insert Table 2 here
280
3.3
Program flow
281
The predictive tool presented in this study considers improvements of the tool
282
developed by Santos et al. [34-35] and Neves et al. [36]. Initially, a window containing
283
a brief description of the predictive tool is displayed, informing the user about the
284
necessary input data (inputs) to perform the simulation in the following steps (see
285
Figure 2).
286
Insert Figure 2 here
287
288
289
As shown in Figure 3, in the next menu the user can select in which industry sector
290
fits the company to be analysed (see Figure 4-a)), including meat, fish, horticultural
291
(fruit and vegetables) and dairy products.
292
Insert Figure 3 here
293
294
295
There are multiple subcategories within each sector. In the specific case of the meat
296
sector industry, there are (1) Slaughterhouse, (2) Sausages industry and (3) Hams
297
industry as shown in Figure 4-b). Other sectors have also specific subcategories.
298
Pressing the button of any of the industry subcategories opens a window requesting
299
the inputs (Figure 4-c)). Here, the user must input the values of the characterization
300
parameters: amount of raw material processed annually [ton], annual electricity
301
consumption [MWh] and the total volume of cold rooms [m3]. Notice that it is always
302
possible to return to the previous menu or close the window using the respective buttons
303
(Back or Exit) for navigation.
304
305
Insert Figure 4 here
306
307
After entering the data according to the SI unit system and considering that the values
308
are within the valid variation domain of each parameter, when the button "Continue" is
309
pressed, the tool starts the computation based on the correlations and displays in a new
310
window the general results that summarize the current state of energy performance of a
311
company. In this window (see Figure 5), plots that relate the characterization parameters
312
of a company are displayed. All plots have a green shaded background that represents
313
the confidence level of 95% of the statistical correlations. In addition, it is also shown in
314
each plot the percent deviation of the parameter under consideration: (Company point –
315
red dot) and Portuguese national average (black dot). If the user wants to analyse a
316
particular plot, pressing the button above the plot will open a new and enlarged window.
317
Insert Figure 5 here
318
319
320
4 Case study
321
4.1
Companies presentation
322
To validate the predictive tool, three companies were selected, hereinafter referred as
323
Company A, Company B and Company C for reasons of business confidentiality, each
324
one for a subcategory of the meat industry: a slaughterhouse, a hams industry and a
325
sausages industry respectively. These companies have not been part of the statistical
326
data used in the development of correlations.
327
The company A is a slaughterhouse with 12 years of activity with 13 workers.
328
According to Portuguese legislation is classified as a micro-enterprise and is dedicated
329
to the meat production, processing annually 1473 tons of raw materials, particularly
330
cattle and pork meat. In its facilities there are nine cold rooms with a total volume of
331
638 m3 and its total covered area is 1117 m2. Some cold rooms are built in sandwich
332
panels and others in masonry insulated by polyurethane. The lighting of the cold rooms
333
is of fluorescent type and the refrigeration tubing are insulated by neoprene as shown in
334
Figure 6 a). The refrigeration compressors have a total rated nominal power of 43 kW
335
distributed evenly by four compressors Bitzer (direct central-circuit). The company
336
annual energy consumption is around 209 MWh. It has capacitors installed for the
337
correction of power factor for an actual value of 0.97. This company also has air-
338
conditioned corridors to avoid losses when accessing the cold rooms as shown in Figure
339
6 b). Additionally, it has a heat recovery unit and a diesel boiler.
340
Insert Figure 6 here
341
342
343
Company B manufactures hams. It has 24 years of experience, employing 15
344
technical workers. It has a covered area of 3000 m2. According to the Portuguese
345
legislation it is classified as a small-enterprise, dedicated to the production of meat,
346
annually processing about 2029.5 tons of feedstock (exclusively pork meat). In its
347
facilities there are 18 cold rooms (freezing, cooling and drying) with a total volume of
348
6668 m3. Eight cold rooms are made of masonry with cork insulation and the remaining
349
ten rooms are built in sandwich panel with polyurethane insulation (see Figure 7). The
350
lighting in the masonry made rooms’ uses incandescent lamps, and the others uses
351
fluorescent lamps. The corridors have no air-conditioning system although the facilities
352
include a heat recovery unit. The compressors of Company B have a total rated power
353
of 221 kW distributed by 18 compressors from the following brands: Bitzer, Copeland
354
and Dorin (individual cooling circuits). The annual global energy consumption is
355
1662.1 MWh and the company has a power factor of 0.91 corrected by capacitors. The
356
company also has a diesel boiler.
357
Insert Figure 7 here
358
359
360
Company C has 15 years of activity, employing only 4 workers. According to the
361
Portuguese legislation is classified as a micro-enterprise, devoted to the production of
362
dry-cured sausages. It processes annually 44.8 tons of raw material, mainly pork meat.
363
The company has three cold rooms with a total volume of 142 m3. However its covered
364
area is 600 m2 since the facilities include rooms for manufacturing the sausages and a
365
smokehouse for drying the products. The cold rooms are made with polyurethane
366
sandwich panels with fluorescent lighting (see Figure 8). The compressors have a total
367
rated power of 10.5 kW distributed by four Bitzer compressors (individual cooling
368
circuits) and the annual global energy consumption is 21.2 MWh. The power factor of
369
the company is 0.86. The facility has not a capacitors system for power factor
370
correction. This particular company does not have a stove either a boiler.
371
Insert Figure 8 here
372
373
374
The details of each industry are gathered in Table 3.
375
Insert Table 3 here
376
377
378
5 Analysis and discussion of results
379
The tool allows predicting the energy performance of a particular company in relation
380
to the Portuguese national average. Depending on the results, the tool also provides
381
practice measures to improve the energy efficiency. Thus, it can be used on energy-
382
decision making process in order to improve the energy efficiency if required and/or
383
desired.
384
This work shows the application of the predictive tool in case studies of industries in
385
the meat sector, particularly, in a slaughterhouse, a hams industry and a sausages
386
industry. The values of the parameters shown in Table 2 (company activity section) for
387
each case study were entered in the predictive tool.
388
Figures 9 to 11 show the overall results of the energy performance of each case study.
389
Based on the predictions, all companies have values of specific electrical energy
390
consumption below average, characterizing them as competitive businesses. In the case
391
of Company A, as shown in Figure 9, note that all plots show a small increment
392
comparing to the national average, such as the energy consumption vs. raw material,
393
where the value is 23% above average (see Figure 9 a). From this plot it appears that for
394
the annual energy consumption of the company, 209 MWh, it would be expected to
395
process more raw material (about 2000 tons). The plot relating the volume of cold
396
rooms vs. raw material predict that cold rooms are capable of storing more 23% of the
397
processed meat annually. An important feature is related to prediction values within the
398
confidence interval of 95% (shaded green). Finally, the plot of specific energy
399
consumption shows that this company has a SEC value 5% below the national average.
400
Therefore it is considered more efficient, having competitive advantage over others with
401
higher energy consumption, and thus being able to practice more competitive prices. If
402
the company processes more raw material, the value of the nominal power of
403
compressors in relation to the amount of raw material processed and volume of cold
404
rooms would be adjusted according to prediction correlation values. The compressors
405
power is 22% higher than expected, which allows concluding that this particular
406
industry can increase the volume of cold rooms or load (raw material) using the
407
installed power. However, based on the audit results there were identified certain issues
408
that may improve the efficiency of this industry, such as the replacement of fluorescent
409
lamps by LED lamps. Furthermore and according to the data collected there is in
410
average one person per hour inside the cold rooms. Despite the air conditioned systems
411
of hallways, this condition increase the thermal load of the cold rooms that lead to an
412
increased energy consumption due to increased working time of compressors to
413
compensate this thermal load. Another negative indicator was the poor condition of the
414
boiler’s insulation. On the other hand the power factor of the company assumes a value
415
close to ideal, being an asset to the energy performance of this industry.
416
Insert Figure 9 here
417
418
419
The company B is the largest from the case studies having a higher number of
420
technical staff than others. At a first glance, it is shown in Figure 10 that most of the
421
prediction values displayed in the plots for this industry are within the confidence
422
interval.
423
424
Insert Figure 10 here
425
426
In the first plot (see Figure 10 a), where energy consumption is related with the raw
427
material processed, it is predicted an energy consumption 31% lower than the average,
428
which allows pointing out that this is a competitive industry. Additionally, the
429
compressors nominal power for the same amount of raw material is 38% lower than the
430
national average. This condition arises from the fact that Company B process more raw
431
material than the national average for the same volume of cold rooms as shown in
432
Figure 10 c). Thus, the specific values related to the raw material processed are below
433
the national average resulting in a lower SEC. Regarding the energy consumption and
434
nominal power of compressors related to the volume of cold rooms (see Figure 10 e)
435
and Figure 10 f)), the values are 7% above and 12% below average, respectively. It
436
should be noted that these latter values lie within the confidence interval, so the
437
uncertainty of the predictions is reduced by the overall analysis of the correlations. The
438
SEC of company B is 32% below the national average being therefore the more efficient
439
company analysed in this case study. The cold storage of raw material is considered as a
440
great advantage in this kind of industry. During the audit, some inefficiencies were
441
detected, namely the lack of insulation in the boiler, the absence of air-conditioning in
442
the hallways and the type of illumination used. Regarding the power factor, the
443
capacitors system ensures that this value remains within the normal parameters.
444
The sausage industry examined within this case study is company C, the smallest one.
445
It is a micro company and the values of the input parameters for predicting the energy
446
performance are in the limit of validity domain of the results.
447
448
Insert Figure 11 here
449
450
Figure 11 a) shown an energy consumption 64% lower than the national average for
451
the amount of the raw material processed. In the remaining plots, the conclusions are
452
similar and all predicted values are below average. The compressors power and the
453
volume of cold rooms are lower than expected by 39% and 23% respectively for the raw
454
material processed when compared to the national average. Figure 11 e) and
455
Figure 11 f) compares the energy consumption and the nominal power of compressors
456
with the volume of cold rooms. In these plots is noted that the predicted values are
457
below the average 86% and 10% respectively. The prediction in Figure 11 e) (86%) is
458
outside the confidence interval due to an amount of raw material processed 23% lower
459
in relation to the volume of the cold room. This condition is verified by the low energy
460
consumption. Considering again Figure 11 a), it is clear that if the company handles
461
more quantity of raw materials, the energy consumption will increase. Finally, by
462
evaluating the plot comparing the SEC with the national average is predicted a value
463
28% lower than the estimated average. It is important to point out that this company
464
uses electricity for heating water. This condition contributes to the low power factor
465
(0.86) which increases the cost of electricity due to the increased reactive power. This
466
problem can be easily solved by installing a small capacitor system. Notice that
467
although the results obtained in the simulations are valid, the value of the amount of raw
468
materials is slightly below the valid domain what can compromise the prediction results,
469
hence the variation between the values of different plots (-86% maximum in Figure 11
470
e) and - 10% minimum in Figure 11 f).
471
472
The predictions results as well as the conclusions obtained from the audits are
473
indicative but allow managers of agrifood companies to be aware about the positioning
474
of their company in relation to the national average in terms of energy performance.
475
However, any change in the company or in the processes require a detailed in-situ study
476
about the energy consumption of different devices that are part of manufacturing and
477
cooling processes, taking into account the various inefficiencies that may exist.
478
Nevertheless, the results obtained by the tool for the prediction of energy performance
479
of meat industries are in line with the failures or inefficiencies detected in the fieldwork.
480
The predictive tool here presented can be adapted to another country or sector, requiring
481
only the modification of the correlations as well as the limit range of application and the
482
confidence intervals.
483
484
6 Conclusions
485
This paper presents a predictive tool of the energy performance of agrifood industry.
486
The main features and workflow of the tool are presented. Three case studies of its
487
application are provided. The selected industries belong each one to a different meat
488
subsector, respectively, slaughterhouse, hams industry and sausages industry. The case
489
studies validate and demonstrate the applicability of the predictive tool. Some
490
conclusions obtained from the graphical analysis of predicted results about the
491
positioning of each company performance in relation to the Portuguese national average
492
are presented. This predictive tool allows performing an evaluation of the energy
493
performance of companies in the agrifood sector comparing the most relevant
494
parameters such as raw materials processed annually, annual energy consumption and
495
volume of cold rooms with the national average. The predictions allows the user to
496
conclude which are the possible weaknesses or strengths of its company. The plot of
497
specific energy consumption is very conclusive since it relates the energy consumption
498
per ton of raw material. It can be seen as a competitiveness indicator. However it should
499
be mentioned that the developed tool only helps managers of agrifood industries in the
500
analysis of the energy performance. It is necessary that the user has sensitivity to
501
identify possible problems of technical origin on the company facilities. Thus, the
502
analysis does not discards the need for a more detailed study to determine particular
503
conditions that can be improved. This conclusion arises from the comparison of results
504
obtained from the simulations and the experimental results from audits.
505
The current state of the predictive tool allows the user to manually enter the data of
506
annual energy consumption, raw material processed annually and volume of cold
507
rooms. With this performance assessment, the user can decide how to improve the
508
energy performance of his company. The practical application of this tool demonstrates
509
its usefulness in helping decision-making in the implementation practice measures for
510
the improvement of energy efficiency. This tool includes also correlations for other
511
sectors of the agrifood sector, such as fish, milk, vine and wine and distribution.
512
Moreover, the predictive tool delivers automatic suggestions for the improvement of
513
energy performance depending on the prediction results. This tool can be used in any
514
country or sector, requiring only the modification of the correlations as well as the limit
515
range of application and the confidence intervals.
516
517
Acknowledgment
518
This work is part of the anchor project "InovEnergy – Eficiência Energética no Sector
519
Agroindustrial" inserted in the InovCluster plan of action. The study was funded by the
520
National Strategic Reference Framework (QREN 2007-2013), POFC (Operational
521
Programme for Competitiveness Factors), 01/SIAC/2011, Ref: 18642.
522
523
References
524
[1] Pachai, A.C.: "From Crandle to table – cooling and freezing of food". in 2nd IIR
525
International Conference on Sustainability and the Cold Chain (ICCC 2013), Paris,
526
France, 2-4 April, 2013.
527
[2] Laguerre, O., Hoang, H. M., Flick, D.: “Experimental investigation and modelling in
528
the food cold chain: Thermal and quality evolution”, Trends in Food Science &
529
Technology 29(2), pp. 87-97, 2013.
530
[3] IEA: 2012 Key World Energy Statistics, International Energy Agency (IEA), 2012.
531
[4] Gogou, E., Katsaros, G., Derens, E., Li, L., Alvarez, G., Taoukis, P.: "Development
532
and applications of the European Cold Chain Database as a tool for cold chain
533
management", in 2nd IIR International Conference on Sustainability and the Cold Chain
534
(ICCC 2013), Paris, France, 2-4 April, 2013.
535
[5] EC: FRISBEE project - Food Refrigeration Innovations for Safety, consumers
536
Benefit, Environmental impact and Energy optimisation along the cold chain in Europe.
537
European Commission (EC) - Research & Innovation.
538
http://ec.europa.eu/research/bioeconomy/food/projects/food_processing/frisbee_en.htm
539
(accessed in 18/02/14)
540
[6] Stahl, V., Alvarez, G., Derens, E., Hoang, H. M., Lintz, A., Hezard, B., El Jabri, M.,
541
Lepage, J. F. Ndoye, F. T., Thuault, D., Gwanpua, S. G., Verboven, P., Geeraerd, A.:
542
"An essential part of the frisbee software tool: Identification and validation of models
543
quantifying the impact of the cold chain on the RTE pork products", in 2nd IIR
544
International Conference on Sustainability and the Cold Chain (ICCC 2013), Paris,
545
France, 2-4 April, 2013.
546
[7] Foster, A., Zilio, C., Corradi, M., Reinholdt, L., Evans, J.: "Freely available cold
547
store energy models", in 2nd IIR International Conference on Sustainability and the
548
Cold Chain (ICCC 2013), Paris, France, 2-4 April, 2013.
549
[8] EC: ICE-E Improving cold storage equipment in europe, European Commission
550
(EC), Energy. http://eaci-projects.eu/iee/page/Page.jsp?op=project_detail&prid=2365
551
(accessed in 18/02/14)
552
[9] Evans, J. A., Hammond, E. C., Gigiel, A. J., Foster, A. M., Reinholdt, L., Fikiin, K.,
553
Zilio, C.: “Assessment of methods to reduce energy consumption of food cold stores”,
554
Applied Thermal Engineering 62(2), pp. 697–705, 2014.
555
[10] Evans, J.A., Foster, A. M., Huet, J.-M., Reinholdt, L., Fikkin, K., Zilio, C.,
556
Houska, M., Landfeld, A., Bond, C., Scheurs, M., T.W.M. van Sambeeck: “Specific
557
energy consumption values for various refrigerated cold stores”, Energy and Buildings
558
74, pp. 141-151, 2014.
559
[11] Eden, M., Colmer, C.: “Improved Cold Chain Management: efficiency and food
560
safety through international project results”, Food Safety and Tecnhology 24(1), pp. 30-
561
32, 2010.
562
[12] EC: CHILL-ON – Enhancing safety at every step on the food supply chain,
563
European
564
http://ec.europa.eu/research/infocentre/article_en.cfm?id=/research/star/index_en.cfm?p
565
=ss-chill-on&item=Infocentre&artid=25333 (accessed in 18/02/14)
Commission
(EC),
Research
&
Innovation.
566
[13] Colmer, C., Kuck, M., Lohmann, M., Bunke, M.: "Novel Technologies to improve
567
safety and transparency of the chilled food supply chain", Innovation in the nordic
568
marine sector, May, 2009.
569
[14] Van der Sluis, S.M.: “Wind energy storage in cold stores: Night Wind”, in Energy
570
Efficiency & Renewable Energy Sources (EE & RES), 4th International Congress for
571
South-East Europe, Sofia, Bulgaria, 7-9 April, 2008.
572
[15] Saint Trofee: Refrigeration Tools, http://www.nightwind.eu/night-wind.html
573
(accessed in 18/02/14)
574
[16] Ge, Y.T., Tassou, S.A.: “Control optimizations for heat recovery from CO2
575
refrigeration systems in supermarket”, Energy Conversion and Management 78, pp.
576
245–252, 2014.
577
[17] Romero, J.A., Navarro-Esbrí, J., Belman-Flores, J.M.: "A simplified black-box
578
model oriented to chilled water temperature control in a variable speed vapour
579
compression system", Applied Thermal Engineering 31(2–3), pp. 329-335, 2011.
580
[18] González-Ramírez, J.E., Leducq, D., Arellano, M., Alvarez, G.: “Energy
581
consumption optimization of a continuous ice cream process“, Energy Conversion and
582
Management 70, pp. 230–238, 2013.
583
[19] Subiantoro, A., Ooi, K.T.: "Economic analysis of the application of expanders in
584
medium scale air-conditioners with conventional refrigerants, R1234yf and CO2",
585
International Journal of Refrigeration 36(5), pp. 1472-1482, 2013.
586
[20] Antonijevic, D.: "An engineering procedure for air side performance evaluation of
587
flat tube heat exchangers with louvered fins", Heat and Mass Transfer 49(1), pp. 117-
588
127, 2013.
589
[21] Khan, M.S., Lee, S., Rangaiah, G.P., Lee, M.: Knowledge based decision making
590
method for the selection of mixed refrigerant systems for energy efficient LNG
591
processes", Applied Energy 111, pp. 1018-1031, 2013.
592
[22] AlQdah, K.: “Potential opportunities for energy savings in a Jordanian poultry
593
company”, Energy Conversion and Management 51, pp. 1651–1655, 2010.
594
[23] Evans, J.A., Hammond, E.C., Gigiel, A.J., Reinholdt, L., Fikiin, K., Zilio, C.:
595
"Improving the energy performance of cold stores", in 2nd IIR International Conference
596
on Sustainability and the Cold Chain (ICCC 2013), Paris, France, 2-4 April, 2013.
597
[24] IAPMEI: PME em números, http://www1.ionline.pt/conteudo/37754-populacao-
598
activa-portuguesa-e-das-que-tem-menos-habilitacoes-literarias- (in Portuguese)
599
[25] Santiago, P., Donaldson, G., Looney A., Nusche, D.: “OECD Reviews of
600
Evaluation and Assessment in Education: Portugal 2012”, Organization for Economic
601
Cooperation Development (OECD) Publishing, 2012.
602
[26] INE: Statistical Yearbook of Portugal, Instituto Nacional de Estatística (INE),
603
Lisbon, Portugal, 2009.
604
[27] Nunes, J.: "Energetic efficiency evaluation in refrigeration systems of agrifood
605
industries in the Beira Interior region", PhD Thesis, University of Beira Interior,
606
Covilhã, Portugal, 2014.
607
[28] Matworks: MATLAB Primer, The Mathworks, Inc., USA, 2013.
608
[29] Matworks: MATLAB Creating Graphical User Interfaces, The Mathworks, Inc.,
609
USA, 2013.
610
[30] Nunes, J., Silva, P.D., Andrade, L.P.: "Energetic efficiency evaluation in
611
refrigeration systems of meat industries", in 23rd International Congress of
612
Refrigeration (ICR 2011), Prague, Czech Republic, August 21 - 26, 2011.
613
[31] Nunes, J., Silva, P.D., Andrade, L.P., Gaspar, P.D.: "Characterization of specific
614
energy consumption of electricity in the Portuguese sausages industry", Energy and
615
Sustainability - WIT Transactions on Ecology and the Environment, vol. 186, 2014.
616
[32] Nunes, J., Silva, P.D., Andrade, L.P., Gaspar, P.D., Domingues, L.C.: "Energetic
617
evaluation of refrigeration systems of horticultural industries in Portugal", in 3rd IIR
618
International Conference on Sustainability and Cold Chain (ICCC 2014), London,
619
United Kingdom, June 23-25, 2014.
620
[33] ASHRAE: Engineering Analysis of Experimental Data - Ashrae Guideline 2,
621
American Society of Heating, Refrigerating and Air-Conditioning Engineers
622
(ASHRAE), Atlanta, 1986.
623
[34] Santos, R., Nunes, J., Silva, P.D., Gaspar, P.D., Andrade, L.P.: "Ferramenta
624
computacional de análise e simulação do desempenho de unidades de conservação de
625
carne através de frio industrial", in VI Congreso Ibérico y IV Congreso Iberoamericano
626
de Ciencias y Técnicas del Frío (CYTEF-2012), Madrid, Spain, February, 2012.
627
[35] Santos, R., Nunes, J., Silva, P.D., Gaspar, P.D., Andrade, L.P.: "Predictive tool for
628
the analysis and simulation of cold room performance in perishable products industry",
629
in 2nd IIR International Conference on Sustainability and the Cold Chain (ICCC 2013),
630
International Institute of Refrigeration (IIR), Paris, France, April 2-4, 2013.
631
[36] Neves, D., Gaspar, P.D., Silva, P.D., Andrade, L.P., Nunes, J.: “Predictive tool for
632
the energy efficiency assessment of cheese industries - Case study of inner region of
633
Portugal”, in V Congreso Iberoamericano de Ciencias y Técnicas del Frío (CYTEF
634
2014), Tarragona, Spain, June 18 - 20 2014.
635
Tables caption:
636
637
Table 1 – Statistical correlation for the energy performance of agrifood companies of the meat
638
sector.
639
Table 2 - Application domain of each parameter of the predictive tool.
640
Table 3 – Companies characterization and activity.
641
642
643
Figures caption:
644
645
646
647
648
Figure 1 – Example of the correlations developed for the sausages industry.
a) Relationship between raw material processed annually and electricity
consumption per year.
b) Relationship between cold room volume and electricity consumption per year.
649
Figure 2 – Cool-OP: Opening window.
650
Figure 3 – Cool-OP menu tree.
651
Figure 4 – Cool-OP: Windows to input the company parameters.
652
a) Menu window.
653
b) Menu window of meat sector.
654
c) Input window of the parameters of the industry (ex: slaughterhouse).
655
Figure 5 – Cool-OP: Window with the overall results (blank).
656
Figure 6 – Company A facilities.
657
a) example of evaporators’ type and location and lighting type of a cold room.
658
b) example of an air-conditioned corridor.
659
Figure 7 – Company B facilities: example of a cold room storing hams.
660
Figure 8 – Company C facilities: example of a cold room storing dry-cured sausages.
661
Figure 9 – Company A: Predictions of energy performance.
662
a) Electrical energy consumption vs Raw material.
663
b) Compressors nominal power vs. Raw material.
664
c) Cold room volume vs. Raw material.
665
d) Specific energy consumption.
666
e) Electrical energy Consumption vs. Cold room volume.
667
f) Compressors nominal power vs. Cold room volume.
668
Figure 10 – Company B: Predictions of energy performance.
669
a) Electrical energy consumption vs Raw material.
670
b) Compressors nominal power vs. Raw material.
671
c) Cold room volume vs. Raw material.
672
d) Specific energy consumption.
673
e) Electrical energy Consumption vs. Cold room volume.
674
f) Compressors nominal power vs. Cold room volume.
675
Figure 11 – Company C: Predictions of energy performance.
676
a) Electrical energy consumption vs Raw material.
677
b) Compressors nominal power vs. Raw material.
678
c) Cold room volume vs. Raw material.
679
d) Specific energy consumption.
680
e) Electrical energy Consumption vs. Cold room volume.
681
f) Compressors nominal power vs. Cold room volume.
682
683
Table 1 – Statistical correlation for the energy performance of agrifood companies of the meat sector.
Industry
Relationships
Electricity consumption, E [MhW] vs. Raw material, RM [ton]
Compressors nominal power, CP [kW] vs. Raw material, RM
[ton]
Cold room volume, V [m3] vs. Raw material, RM [ton]
Slaughterhouses
Sausages
Hams
E = 0.2041 RM - 138.97
E = 0.2376 MP + 24.18
E = 0.9945 RM + 159.93
(R2 = 0.57)
(R2 = 0.92)
(R2 = 0.86)
CP = 0.0520 RM - 51.46
CP = 0.0702 RM + 11.49
CP = 0.1082 RM + 86.20
(R2 = 0.71)
(R2 = 0.68)
(R2 = 0.73)
V = 0.5151 RM - 265.58
V = 4.3795 RM - 87.42
V = 3.8629 RM + 785.44
(R2 = 0.90)
(R2 = 0.82)
(R2 = 0.89)
E = 0.4557 V - 112.71
E = 0.3243 V - 6.84
E = 0.2667 V – 228.61
(R2 = 0.84)
(R2 = 0.96)
(R2 = 0.96)
3
Electricity Consumption, E [MWh] vs. Cold room volume, V [m ]
684
685
CP = 0.1102 V - 36.72
CP = 0.0572 V + 3.41
CP = 0.0323 V + 32.05
Compressors nominal power, CP [kW] vs. Cold room volume, V
[m3]
(R2 = 0.94)
(R2 = 0.93)
(R2 = 0.90)
Average specific energy consumption, SEC [kWh ton-1]
148.5
660.2
1208.0
Table 2 – Application domain of each parameter of the predictive tool.
686
687
688
Raw Material
Volume of cold rooms
Compressors nominal power
[ton]
[m3]
[kW]
Slaughterhouses
1000-5000
500-3000
25-300
Sausages
50-1100
10-1000
10-100
Hams
50-2500
1500-11000
80-350
Industry
689
Table 3 – Companies characterization and activity.
690
Company activity
Company
Company A
Company B
Company C
Slaughterhouses
Hams industry
Sausages industry
Quantity of raw material [ton]
1473.0
2029.5
44.8
Electricity consumption [MWh]
209.0
1662.1
21.2
Total volume of cold rooms [m3]
638
6668
142
Compressors nominal power [kW]
43
221
10.5
Company age
12
24
15
Number of workers
13
15
4
Micro company
Small company
Micro company
9
18
3
1117
3000
600
Fluorescent
Fluorescent/Incandescent
Fluorescent
Neoprene
Neoprene
Polyurethane
Polyurethane
Polyurethane/cork
Polyurethane
Capacitors battery
Yes
Yes
No
Power factor
0.97
0.91
0.86
Boiler
Yes
Yes
No
Air-conditioned corridors
Yes
No
No
Meat subsector
Company characterization
Company size
691
Number of cold rooms
Covered area [m2]
Lighting
Tube insulation
Chambers insulation
300
Electricity consumption [MWh]
Electricity consumption [MWh]
300
250
E = 0.2376 RM + 24.18
R² = 0,92
200
150
100
50
0
0
200
400
600
800
Raw material [ton]
1000
1200
250
E = 0.3243 V - 6,84
R² = 0.96
200
150
100
50
0
0
200
400
600
Volume [m3]
b) Relationship between cold room volume and electricity consumption
consumption per year.
per year.
693
694
696
1000
a) Relationship between raw material processed annually and electricity
692
695
800
Figure 1 – Example of the correlations developed for the sausages industry.
697
698
699
700
701
702
Figure 2 – Cool-OP: Opening window.
Figure 3 – Cool-OP menu tree.
c) Input window of the parameters
b) Menu window of
a) Menu window.
of the industry (ex:
meat sector.
slaughterhouse).
Figure 4 – Cool-OP: Windows to input the company parameters.
Figure 5 – Cool-OP: Window with the overall results (blank).
a) example of evaporators’ type and location and lighting type of a cold
room.
Figure 6 – Company A facilities:
b) example of an air-conditioned corridor.
Figure 7 – Company B facilities: example of a cold room storing hams.
Figure 8 – Company C facilities: example of a cold room storing dry-cured sausages.
a) Electrical energy consumption vs. Raw material.
d) Specific energy consumption.
b) Compressors nominal power vs. Raw material.
c) Cold room volume vs. Raw material.
e) Electrical energy Consumption vs. Cold room
volume.
Figure 9 – Company A: Predictions of energy performance.
f) Compressors nominal power vs. Cold room
volume.
a) Electrical energy consumption vs. Raw material.
d) Specific energy consumption.
b) Compressors nominal power vs. Raw material.
c) Cold room volume vs. Raw material.
e) Electrical energy Consumption vs. Cold room
f) Compressors nominal power vs. Cold room
volume.
volume.
Figure 10 – Company B: Predictions of energy performance.
a) Electrical energy consumption vs. Raw material.
d) Specific energy consumption.
b) Compressors nominal power vs. Raw material.
c) Cold room volume vs. Raw material.
e) Electrical energy Consumption vs. Cold room
f) Compressors nominal power vs. Cold room
volume.
volume.
Figure 11 – Company C: Predictions of energy performance.