Environmental Project, 951 – Comparisons of Energy Consumption for Refrigeration in Supermarkets

Summary

The aim of the project presented in this report has been to compare the energy consumption for refrigeration in four supermarkets of approximately the same size but with different layout of the refrigeration system. Two of the supermarkets have newly developed cascade refrigeration systems with CO₂ in the low-temperature circuit and R404A in the high-temperature circuit. All the display cases were cooled by CO₂ with dry-expansion evaporators in the freezers and flooded evaporators in the refrigerators. The other two supermarkets have conventional multiplex refrigeration systems with R404A as refrigerant and dry-expansion evaporators in all display cases. The energy consumption was measured during a period of five month (1/8 – 31/12 2003). The energy consumption of the four systems cannot be compared directly because the systems were not the same size and their condensation temperatures were different due to heat recovery. Therefore, a reference supermarket is used, where the energy consumption is based on the measured energy consumptions corrected to the same reference condensations temperature.

Two models have been used to estimate the corrected energy consumption. The first model is based on a simple scaling factor calculated as the ratio between the nominal cooling capacity of the reference supermarket and the nominal cooling capacity of the actual supermarket. The nominal cooling capacity is calculated from the specification of the cooling needs for the individual display cases. The second model uses a prescribed load profile for the low and high temperature circuits of the refrigeration system. The load profile used is based on the measured load profile of one of the supermarkets during the period of measurement. The measured load profile is expressed as the ratio between the measured load and the highest load in the period of measurement. For the reference system the maximum high temperature load is set at 110 kW and the maximum low temperature load is set at 40 kW.

The mean COP (the ratio between the cooling load and the power supplied) has been calculated for the whole period of measurement.

According to the first model (scaling factor), the energy consumption of the cascade plants, including the pumping power for the flooded evaporators, is 2/3 of the energy consumption of the conventional multiplex system. This model has later been rejected because the calculated cooling load for the reference supermarket depends on which supermarket is used as a basis for the calculation, whereas only the power consumption of the reference supermarket should depend on which supermarket is used as a basis. The differences in cooling loads seem to be caused by a very high dependency of the nominal data.

According to the second model (load profile), the calculated energy consumptions of the two cascade systems and one of the multiplex systems were the same within the degree of measuring accuracy. The last multiplex system has a much higher energy consumption and the reason for this must be a lower efficiency of the compressors used.

For the cascade systems the energy consumption of the circulation pump for the flooded evaporators is approximately 10% of the total energy consumption. It is possible to reduce the energy consumption of the circulation pump by adjusting the capacity of the pump to the needs. At present the pump is running at full capacity independently of the actual need.

The over-all conclusion from the comparisons is that the new cascade systems have an energy consumption equal the energy consumption of a well dimensioned conventional refrigeration system and that it is possible to lower the energy consumption of the cascade system by implementing a control strategy for the circulation pump.