Entropy analysis in data center chiller operation

Abstract

Background: Refrigeration system s are widely used in various industries. To achieve temperatures below the ambient temperature and remove excess heat, it is required to expend work (power). Such expenses often constitute a significant portion of energy consumed by the entire facility.

For example, refrigeration equipment used in a most fast-growing industry, construction of data centers (DCs), consume up to 40% of the total power consumption [1].

To reduce energy consumption, it is important to analyze the operation at all life cycle stages from design to operation. Theoretical results often lag far behind practice and require changing both the refrigeration system design and its control algorithms.

This paper presents an analysis of a data center liquid refrigeration unit and the actions aimed at improving the system performance.

Aim: To demonstrate the requirement in and possible uses of operational analysis of refrigeration system at the commissioning stage.

Methods: Chiller losses in the data center refrigeration system were analyzed using the entropy-statistical thermodynamic analysis [2]. Based on the analysis, actions were taken to improve the chiller performance.

Results: Entropy-statistical thermodynamic analysis at life cycle stages of a refrigeration unit allows to identify components that require intervention (adjustment, replacement, etc.). Deviation of the effective refrigeration coefficient from the design value was 13.67% after commissioning and 4.14% after evaporator replacement. As the design specification of the refrigeration unit indicated that the refrigeration coefficient not be lower than 5.4, the achieved value of 5.34 with 1.04% deviation from the specification is acceptable.

Conclusion: Entropy-statistical thermodynamic analysis at the design and commissioning stages allow to achieve the specified performance values. In addition, the distribution of losses allowed to identify a component with the highest losses, which was then adjusted.

Keywords

entropy-statistical analysis / performance / data center cooling / chiller

References

[1] Chichyants AE. Improving energy efficiency of data center air conditioning systems. Vestnik Nauki. 2023; 3(5(62)):853–859. (In Russ.)

[2] Arkharov AM. Fundamentals of cryology. Entropy-statistical analysis of low-temperature systems. Moscow: MGTU im NE Baumana; 2014. (In Russ.)

[3] Arkharov AM, Shishov VV. Entropy-statistical analysis of low-temperature refrigeration cycles. Refrigeration Technology. 2014;8:50–53. (In Russ.)

[4] Arkharov AM, Shishov VV, Talyzin MS. Entropy-statistical analysis of low-temperature refrigeration cycles and choosing of optimal refrigeration system for retail. Refrigeration Technology. 2016;3:30–34. (In Russ.)

[5] Talyzin MS, Skolov AV. Selection of refrigerant for use in chillers. Refrigeration Technology. 2024;113(1):13–20. (In Russ.) doi: 10.17816/RF632560

[6] Shishov VV, Talyzin MS. Practical application of the entropic and statistical method of the analysis of refrigeration cycles. Refrigeration Technology. 2015;3:25–28. (In Russ.)

RIGHTS & PERMISSIONS

Eco-Vector