A PSO-LSTM Framework for Optimizing Temperature and Humidity in Data Centers to Minimize Energy Consumption

With the rapid development of data centers, optimizing temperature and humidity control has become crucial for reducing energy consumption and improving operational efficiency. This paper proposes a PSO-LSTM-based optimization framework for determining the optimal temperature and humidity settings in data centers to minimize total energy consumption. The framework employs Particle Swarm Optimization (PSO) to determine the optimal air temperature that balances server and cooling power consumption. Additionally, a Long Short-Term Memory (LSTM) network is trained to predict the optimal humidity level, ensuring stable system performance. The proposed method is evaluated through simulations under varying environmental conditions, demonstrating its effectiveness in reducing energy consumption while maintaining data center operational reliability. The results indicate that the optimized temperature and humidity settings can significantly enhance energy efficiency and reduce overall power usage.