An Intelligent Thermal Management Strategy for a Data Center Prototype Based on Digital Twin Technology
Source abstract
Abstract
Data centers contribute to roughly 1% of global energy consumption and 0.3% of worldwide carbon dioxide emissions. The cooling system alone constitutes a substantial 50% of total energy consumption for data centers. Lowering Power Usage Effectiveness (PUE) of data center cooling systems from 2.2 to 1.4, or even below, is one of the critical issues in this thermal management area. In this work, a digital twin system of an Intelligent Data Center (IDC) prototype is designed to be capable of real-time monitoring the temperature distribution. Moreover, aiming to lower PUE, Deep Q-Learning Network (DQN) is further established to make optimization decisions of thermal management during cooling down of the local hotspot. The entire process of thermal management for IDC can be real-time visualized in Unity, forming the virtual entity of data center prototype, which provides an intelligent solution for sustainable data center operation.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Digital Transformation in Industry
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Aeronautics Science and Engineering, Beihang University, Beijing 100191, China
China Communications Information & Technology Group Co., Ltd., Beijing 101399, China
School of General Engineering, Beihang University, Beijing 100191, China
Institute of Thermodynamics, Technical University of Munich, 80333 Munich, Germany
Funding
China Communications Information & Technology Group Co., Ltd.
References (37)
Show source-provided references
1. Recalibrating global data center energy-use estimates10.1126/science.aba3758 ↗
2. How to stop data centres from gobbling up the world’s electricity10.1038/d41586-018-06610-y ↗
3. Development and experimental study of an independent row-based cooling system for improving thermal performance of a data center10.1016/j.applthermaleng.2019.114857 ↗
4. Data center energy consumption modeling: A survey10.1109/COMST.2015.2481183 ↗
5. Cooling energy consumption investigation of data center it room with vertical placed server10.1016/j.egypro.2017.03.581 ↗
6. Data center power minimization with placement optimization of liquid-cooled servers and free air cooling
7. Shehabi, A., Smith, S., Sartor, D., Brown, R., Herrlin, M., Koomey, J., Masanet, E., Horner, N., Azevedo, I., and Lintner, W. (2016). United States Data Center Energy Usage Report, Lawrence Berkeley National Laboratory.10.2172/1372902 ↗
8. Reinforcement learning for green and reliable data center
9. Comparison of data driven modeling approaches for temperature prediction in data centers10.1016/j.ijheatmasstransfer.2019.02.041 ↗
10. Physics-informed neural networks (PINNs) for fluid mechanics: A review10.1007/s10409-021-01148-1 ↗
11. Physics-informed neural networks for heat transfer problems10.1115/1.4050542 ↗
12. Physics-informed deep learning for incompressible laminar flows10.1016/j.taml.2020.01.039 ↗
13. Physics guided machine learning using simplified theories10.1063/5.0038929 ↗
14. Ribeiro, M.D., Rehman, A., Ahmed, S., and Dengel, A. (2020). DeepCFD: Efficient steady-state laminar flow approximation with deep convolutional neural networks. arXiv.
15. Machine learning–accelerated computational fluid dynamics10.1073/pnas.2101784118 ↗
16. A point-cloud deep learning framework for prediction of fluid flow fields on irregular geometries10.1063/5.0033376 ↗
17. TransCFD: A transformer-based decoder for flow field prediction10.1016/j.engappai.2023.106340 ↗
18. Deep learning-based heat optimization techniques for forecasting indoor temperature changes10.1007/s41324-023-00546-w ↗
19. Deep reinforcement learning to optimise indoor temperature control and heating energy consumption in buildings10.1016/j.enbuild.2020.110225 ↗
20. Explainable AI for engineering design: A unified approach of systems engineering and component-based deep learning demonstrated by energy-efficient building design10.1016/j.aei.2024.102843 ↗
21. Interpreting the neural network model for HVAC system energy data mining10.1016/j.buildenv.2021.108449 ↗
22. Human-level control through deep reinforcement learning10.1038/nature14236 ↗
23. Zhang, Z., Zeng, Y., Liu, H., Zhao, C., Wang, F., and Chen, Y. (2022, January 25–29). Smart DC: An AI and digital twin-based energy-saving solution for data centers. Proceedings of the NOMS 2022-2022 IEEE/IFIP Network Operations and Management Symposium, Budapest, Hungary.10.1109/NOMS54207.2022.9789853 ↗
24. Transforming cooling optimization for green data center via deep reinforcement learning10.1109/TCYB.2019.2927410 ↗
25. Grishina, A., Chinnici, M., Kor, A.-L., Rondeau, E., and Georges, J.-P. (2020). A machine learning solution for data center thermal characteristics analysis. Energies, 13.10.20944/preprints202007.0325.v1 ↗
26. Green Data Center Cooling Control via Physics-Guided Safe Reinforcement Learning
27. Optimizing data center energy efficiency via event-driven deep reinforcement learning10.1109/TSC.2022.3157145 ↗
28. Reinforcement learning control for data center refrigeration systems
29. Mastering the game of Go with deep neural networks and tree search10.1038/nature16961 ↗
30. Global optimization strategy of prosumer data center system operation based on multi-agent deep reinforcement learning10.1016/j.jobe.2024.109519 ↗
31. Wang, Y., Sun, Y., Cheng, B., Jiang, G., and Zhou, H. (2023, January 20–22). DQN-Based Chiller Energy Consumption Optimization in IoT-Enabled Data Center. Proceedings of the 2023 IEEE 23rd International Conference on Communication Technology (ICCT), Wuxi, China.10.1109/ICCT59356.2023.10419683 ↗
32. DDQN-based data laboratory energy consumption control model10.1504/IJSNET.2024.137338 ↗
33. A digital-twin and machine-learning framework for precise heat and energy management of data-centers10.1007/s00466-022-02152-3 ↗
34. Digital twin-driven energy consumption management of integrated heat pipe cooling system for a data center10.1016/j.apenergy.2024.123840 ↗
35. Piras, G., Agostinelli, S., and Muzi, F. (2025). Smart Buildings and Digital Twin to Monitoring the Efficiency and Wellness of Working Environments: A Case Study on IoT Integration and Data-Driven Management. Appl. Sci., 15.10.3390/app15094939 ↗
36. Digital twin-driven product design, manufacturing and service with big data10.1007/s00170-017-0233-1 ↗
37. Malone, C., and Belady, C. (2006, January 18–20). Metrics to characterize data center & IT equipment energy use. Proceedings of the Digital Power Forum, Richardson, TX, USA.