Optimization of a free cooling system integrated with cold thermal energy storage in data center based on model predictive control

With the rapid development of information technology, energy consumption in data centers has become increasingly prominent. As a core component, cooling systems account for substantial energy use while offering significant energy-saving potential, making them crucial for energy efficiency optimization. To address energy conservation in cooling systems, a free cooling system integrated with cold thermal energy storage is investigated in this study. Using typical meteorological parameters of Wuhan as a case study, a genetic algorithm (GA)-based model predictive control (MPC) strategy is employed to optimize system performance, and its adaptability across different climatic zones in China is evaluated. The results demonstrate that optimizing with power usage effectiveness (PUE) minimization as the objective function reduces the PUE value by 0.018 compared to the baseline system. When applied nationwide, lower PUE values are observed in regions with more abundant free cooling resources. After MPC optimization, the most significant improvements are exhibited in the mild climate zone, where a maximum PUE reduction of 0.0185 is achieved compared to pre-optimized systems.