A data center water cooling unit energy consumption optimization method and system based on DQN
By using a DQN-based energy consumption optimization method for water-cooled units, and leveraging IoT data and neural network models, the cooling temperature of the water-cooling system is optimized. This solves the generalization and stability problems of traditional methods, and effectively reduces data center energy consumption and optimizes PUE.
Patent Information
- Application Number
- CN202311239197.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-25
- Publication Date
- 2026-09-11
- Estimated Expiration
- 2043-09-25
AI Technical Summary
Traditional energy consumption optimization methods for data center water cooling systems face challenges in terms of generalization ability, stability, and timeliness, making it difficult to effectively reduce PUE.
A water-cooled unit energy consumption optimization method based on deep reinforcement learning (DQN) is adopted. The device measurement data is obtained through the Internet of Things, a neural network model is constructed, and the model is modeled as a Markov decision process to optimize the cooling water outlet temperature (CWOT) to reduce energy consumption.
It significantly reduces the energy consumption of water-cooled units, optimizes the PUE of data centers, and features fast algorithm convergence, resulting in better optimization performance and greater flexibility.