arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.15085cs.AIcs.LGcs.SYeess.SY
收藏

基于离线强化学习的数据中心冷却系统优化

Data Center Cooling System Optimization Using Offline Reinforcement Learning

Institute for AI Industry Research, Tsinghua University(清华大学人工智能产业研究院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Global Data Solutions Co., Ltd.(全球数据解决方案有限公司)

Xianyuan Zhan, Xiangyu Zhu, Peng Cheng, Xiao Hu, Ziteng He, Hanfei Geng, Jichao Leng, Huiwen Zheng, Chenhui Liu, Tianshun Hong, Yan Liang, Yunxin Liu, Feng Zhao

AI总结:

针对数据中心冷却系统节能优化的挑战,提出物理信息离线RL框架,通过符合时间反转对称性的图神经网络建模,在生产数据中心实验中实现14~21%节能且满足安全约束,验证其在工业控制中的潜力。

摘要:

信息技术和人工智能的最新进展推动了全球数据中心(DC)行业的快速扩张,同时也带来了对数据中心电力的巨大需求。在典型的数据中心中,约30~40%的能源用于冷却系统而非计算机服务器,这迫切需要开发新的节能优化技术来优化数据中心冷却系统。然而,优化此类现实工业系统面临诸多挑战,包括但不限于缺乏可靠的仿真环境、历史数据有限以及严格的安全和控制鲁棒性要求。在这项工作中,我们提出了一种新颖的物理信息离线强化学习(RL)框架,用于数据中心冷却系统的能效优化。所提出的框架使用专门设计的图神经网络架构对服务器机房内的复杂动态模式和物理依赖关系进行建模,该架构符合基本的时间反转对称性。由于其性能良好且可泛化的状态-动作表示,该模型能够利用有限的实际运行数据进行样本高效且鲁棒的 latent space 离线策略学习。我们的框架已成功部署并在大型生产数据中心中验证了其对空气冷却单元(ACUs)的闭环控制。我们在生产数据中心环境中进行了总共2000小时的短期和长期实验。结果表明,我们的方法在数据中心冷却系统中实现了14~21%的节能,且未违反任何安全或操作约束。我们的结果证明了离线强化学习在解决广泛的数据有限、安全关键的现实工业控制问题方面的巨大潜力。

英文摘要:

The recent advances in information technology and artificial intelligence have fueled a rapid expansion of the data center (DC) industry worldwide, accompanied by an immense appetite for electricity to power the DCs. In a typical DC, around 30~40% of the energy is spent on the cooling system rather than on computer servers, posing a pressing need for developing new energy-saving optimization technologies for DC cooling systems. However, optimizing such real-world industrial systems faces numerous challenges, including but not limited to a lack of reliable simulation environments, limited historical data, and stringent safety and control robustness requirements. In this work, we present a novel physics-informed offline reinforcement learning (RL) framework for energy efficiency optimization of DC cooling systems. The proposed framework models the complex dynamical patterns and physical dependencies inside a server room using a purposely designed graph neural network architecture that is compliant with the fundamental time-reversal symmetry. Because of its well-behaved and generalizable state-action representations, the model enables sample-efficient and robust latent space offline policy learning using limited real-world operational data. Our framework has been successfully deployed and verified in a large-scale production DC for closed-loop control of its air-cooling units (ACUs). We conducted a total of 2000 hours of short and long-term experiments in the production DC environment. The results show that our method achieves 14~21% energy savings in the DC cooling system, without any violation of the safety or operational constraints. Our results have demonstrated the significant potential of offline RL in solving a broad range of data-limited, safety-critical real-world industrial control problems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑