Study on Control and Optimization of Data Center Chiller System Based on Phase Change Cold Storage

This study addresses the inefficiency of cooling systems in data centers by developing an air conditioning chilled water system incorporating phase change thermal energy storage. A comprehensive control framework is established through the construction of energy consumption models for pumps, chillers, and cooling towers, as well as a dynamic cooling capacity output model for the phase change storage unit. To navigate the complexity of multi-variable optimization, a hybrid algorithm combining Simulated Annealing and Particle Swarm Optimization is proposed, enhancing convergence and mitigating the risk of local optima. Experimental outcomes indicate that the optimized system reduces operational costs by approximately 30 percent and decreases the Power Usage Effectiveness from 1.5 to 1.38, significantly enhancing the energy efficiency of data centers.