Ice-E Cold Store Survey: 2.1. Detailed Survey Tool Development of Survey Tool
Ice-E Cold Store Survey: 2.1. Detailed Survey Tool Development of Survey Tool
1. INTRODUCTION
The cold chain is believed to be responsible for approximately 2.5% of global greenhouse gas (GHG)
emissions through direct and indirect (energy consumption) effects. Studies have shown that leakage of
refrigerants may be higher than 17% in industrial plant (Clodic and Palandre 2004). Refrigeration energy
consumed in the food chain has not been accurately quantified due to a lack of measured energy
consumption and process throughput data in most sectors (Swain, 2006). Overall figures would indicate that
excluding domestic refrigeration, approximately 50% of the energy is associated with retail and commercial
refrigeration and 50% with chilling, freezing and storage (Market Transformation Programme).
Cold storage rooms consume considerable amounts of energy. Within cold storage facilities 60-70% of the
electrical energy used is for refrigeration. Therefore cold store users have considerable incentive to reduce
energy consumption. In Europe there are 60-70 million cubic meters of cold storage for food. In 2002 the IIR
estimated that cold stores used between 30 and 50 kWh/m3/year (Duiven and Binard, 2002). Previous surveys
carried out on a small number of cold stores have shown that energy consumption can dramatically exceed
this figure, often by at least double (Evans and Gigiel, 2007, 2010). These surveys also demonstrated that
energy savings of 30-40% were achievable by optimising usage of the stores, repairing current equipment
and by retrofitting of energy efficient equipment. However, cold store operators are often reluctant to install
new equipment without sufficient information on savings that could be achieved.
There are few published surveys comparing the performance of more than a few cold stores. The most
comprehensive recent survey was carried out in New Zealand by Werner et al (2006) which compared
performance of 34 cold stores. This demonstrated that there was a large variation in energy consumed by
cold stores and that savings of between 15 and 26% could be achieved by applying best practice
technologies.
The performance of European cold stores has never been compared in detail and there is little information to
compare their performance with other stores Worldwide. With government targets to reduce energy and
reduce emissions of greenhouse gasses the need to benchmark and understand potential energy and GHG
reductions is of great interested to end users. To enable end users to improve the performance of their cold
stores a project called ‘Improving Cold storage Equipment in Europe’ (ICE-E) was developed with 8
partners from across Europe. The initial aim of the project was to collect data to benchmark the performance
of cold stores in Europe.
As part of the ICE-E project, two internet based surveys were developed and data collected to determine
energy usage in different cold store types, sizes and configurations. Results from these surveys are presented
in this report and the data analysed to determine whether there were any common factors that affected
performance of the cold stores.
Data collected
The survey parameters collected are shown in Table 1. In all cases the users were asked to rate the accuracy
of the data they submitted. The collected data was retained on a server where users could return to update
information or add further data.
Benchmark analysis
Once users had input data they could then compare the performance of their store through an automatic
benchmark analysis. This enabled them to compare the energy used by their cold store system with systems
of a similar size and product throughput (Figure 2). In addition users could compare the set point
temperatures, food type and room function and refrigerant type with others in the survey. In all comparisons
the user had the ability to define the range over which comparisons were carried out.
Cold store 3
Data collected
The data collection form is shown in Figure 3. A limited data set was collected (set temperature, area and
volume of the store, food throughput and energy usage per year) which reflected what were considered to be
the most important factors affecting energy use in cold stores.
Benchmark analysis
Once data was submitted the information was input manually into the main benchmark survey and
information sent directly to the cold store operator.
3. RESULTS
Belgium
Bulgaria
1 China
15 8 4
13 Czech Republic
4
11 1 Denmark
5
2 France
Germany
7 1 Greece
12 Ireland
Italy
Mexico
129 Netherlands
New Zealand
Portugal
Romania
69 Serbia
Spain
Sweden
4 5 Switzerland
1 1 1 United Kingdom
1
US
100
80
60
40
20
0
Chilled Frozen Mixed
Figure 5. Average SEC (Specific Energy Consumption) and standard deviation around means for chilled,
frozen and mixed cold stores.
3.3. Country
Large variations in SEC were shown between countries. Significant differences were found between chilled
and frozen stores but not mixed stores in the countries within the survey. Figure 6 shows average SEC for
chilled, frozen and mixed stores in each country where data was collected and the standard deviations around
the means. This showed large variability in the SEC between countries and within countries. Due to the
limited number of data sets for some countries it would not have been possible to analyse data from each
country separately. All further analysis was carried out on data divided into chilled, frozen and mixed stores.
200
150
100
50
Figure 6. Average SEC per country and standard deviation around means where replicate data was available
(where no s.d. is plotted, this is because there is only one data point).
3.4. Relationship between energy use and store size
The relationship between store energy consumption (in kWh/year) and the information collected was
investigated using multiple regression. As part of this analysis the data was found to be near to a normal
distribution.
16,000,000
14,000,000
12,000,000
10,000,000
8,000,000
6,000,000
4,000,000
2,000,000
0
0 50,000 100,000 150,000 200,000 250,000
Volume (m 3)
Figure 7. Relationship between store volume and total energy use per year (kWh/year) for chilled stores.
14,000,000
y = 39.576x + 859987
R² = 0.5632
12,000,000
10,000,000
8,000,000
6,000,000
4,000,000
2,000,000
0
0 50,000 100,000 150,000 200,000 250,000 300,000 350,000
Volume (m 3)
Figure 8. Relationship between store volume and total energy use per year (kWh/year) for frozen stores
(linear regression).
14,000,000
12,000,000
y = 374.93x 0.8173
R² = 0.6567
10,000,000
8,000,000
6,000,000
4,000,000
2,000,000
0
0 50,000 100,000 150,000 200,000 250,000 300,000 350,000
Volume (m 3)
Figure 9. Relationship between store volume and total energy use per year (kWh/year) for frozen stores (non
linear regression).
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000
Volume (m 3)
Figure 10. Relationship between store volume and total energy use per year (kWh/year) for mixed stores
(linear regression).
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000
Volume (m 3)
Figure 11. Relationship between store volume and total energy use per year (kWh/year) for mixed stores
(non linear regression).
3.8. All stores
A comparison of the best fit regressions for chilled, frozen and mixed stores is shown in Figure 12. It is
interesting to note that mixed and frozen stores had a relatively similar relationship between volume and
annual energy (although statistically the regressions lines were significantly different at P<0.01). At volumes
below 22,000 m3 chilled store used less energy than frozen or mixed stores but at volumes above 22,000 m3
chilled stores used more energy than frozen or mixed stores. This was mainly due to a cluster of smaller
chilled stores that had low energy consumption.
kWh/year
Chilled Frozen Mixed 331 cold stores (mixed chilled and frozen)
20,000,000
18,000,000
16,000,000
14,000,000
12,000,000
10,000,000
8,000,000
6,000,000
4,000,000
2,000,000
0
0 50,000 100,000 150,000 200,000 250,000 300,000
Volume (m3)
Figure 12. Comparison between store volume and energy use per year (kWh/year) for all stores.
The SEC for the cold stores examined varied considerable. Data from all stores and for all stores with the
10% and 20% upper and lower values removed are shown in Table 2. The distribution of the SEC values for
chilled, frozen and mixed cold stores are shown in Figure 13.
Table 2. Range in SEC values for cold stores examined.
% of population
20%
Chilled Frozen Mixed
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
4. CONCLUSIONS
The data collected in the ICE-E survey showed that there was large variability in the energy used by cold
stores. The SEC varied between 4 and 250 kWh/m3/year for chillers, between 6 and 240 kWh/m3/year for
freezers and between 23 and 157 kWh/m3/year for mixed stores. Duiven and Binard (2002) estimated that
cold stores should use between 30 and 50 kWh/m3/year.. The data collected in this survey demonstrated that
47% of chilled stores, 35% of frozen stores and 50% f mixed stores had an SEC of less than 50 kWh/m3/year.
This demonstrates that there is considerable potential to reduce energy consumption in cold stores.
Large differences were found between cold stores in different countries but the reasons for this were not able
to be extracted due to lack of replicate data.
Differences were found between chilled, frozen and mixed usage cold stores. The major influence on annual
energy consumption in all store types was the volume of the store. For chilled stores volume accounted for
93% of the variation in annual energy consumption. In frozen and mixed stores, volume accounted for 66-
76% of the variation in annual energy consumption but this was a non linear relationship. In frozen and
mixed stores other factors had minimal impact on variation in annual energy consumption. This may have
been due to lack of replicate data for some factors recorded. Further work is required to examine the impact
of some of these factors. There also appears, especially with frozen and mixed stores, a factor or factors
affecting annual energy consumption that were not collected in the survey. The survey was not able to assess
heat loads from chamber freezing or processing in chamber (although this information was requested it was
rarely provided). Therefore the variations in energy consumption may be explained by the additional heat
loads in some chambers. This requires further investigation.
The performance of all stores (chilled, frozen and mixed) was statistically different. However, there was
more relationship between the performance of frozen and mixed stores than there was between chilled and
frozen or chilled and mixed stores. The energy used by chilled stores was less than frozen or mixed stores at
volumes below 22,000 m3 but was higher above this value. This might indicate that large frozen stores tend
to be long term stores with less usage and that larger chilled stores have high usage (e.g. large regional
distribution centres). This again requires further investigation.
It would be expected that larger stores would be more efficient and have a lower SEC than smaller stores.
The indications were that this was only the case for frozen and mixed stores. For chilled stores the
relationship between volume and store size was linear. This indicates that chilled and frozen/mixed stores are
affected by transmission heat loads in different ways. It is possible that transmission is more dominant in
stores with lower temperatures and so the impact of surface to volume ratio (which is less in a larger store) is
greater than in chilled stores. It might also be expected that usage of chilled stores may be greater than that of
frozen stores (more movement of food, more door openings etc) and that this may be a more dominant factor
affecting energy than transmission. However, there was no relationship between annual energy consumption
and food throughout in chilled stores and so this does not appear to be the answer.
The analysis demonstrated a surprising lack of relationships between the factors recorded (apart from
volume) and annual energy consumption. There was for example no relationship for any store types with
temperature of the store even though the range in temperatures recorded were relatively wide ranging (13°C
for chilled and 5°C for frozen) and there was an extensive data set. In other instances the lack of any
relationship may have been due to the restricted data sets available. It would therefore be useful to collect
further data on the factors that were indicated to be important by the regression analysis.
The data collected provides an indication of the factors that most affect the energy used by cold stores. This
provides a useful framework to develop labelling of cold stores and the factors that should be considered
when creating a benchmarking or labelling scheme.
5. REFERENCES
Clodic D, Palandre L. 2004. Determination of Comparative. HCFC and HFC Emission Profiles for the Foam
and Refrigeration Sectors until 2015. Part 1. Refrigerant Emission. Profiles. Centre d’Energetique.
Duiven JE, Binard P. 2002. Refrigerated storage: new developments. Bulletin of the IIR – No. 2002-2.
Evans, JA, Gigiel AJ. 2007. Reducing the energy consumption in cold stores. The 22nd IIR International
Congress of Refrigeration. Beijing, China. August 21-26, 2007.
Evans JA, Gigiel A. 2010. Reducing energy consumption in cold storage rooms. IIR ICCC, Cambridge 29-
31 March 2010.
Market Transformation Programme – http://www.mtprog.com/
Swain MJ. 2006. Improving the energy efficiency of food refrigeration operations. IChemE Food and Drink
Newsletter, 4 Sept.
Werner SRL, Vaino F, Merts I, Cleland DJ. 2006. Energy use in the New Zealand cold storage industry. IIR-
IRHACE Conference, The University of Auckland, 2006.