Simulation-based Multi-unit Collaborative Optimization in Data Centers

Abstract With the rapid development of information technology in China, problem energy conservation data centers has become increasingly prominent, and green center model is inevitable. The refrigeration equipment which main power consumption affected by environment, room temperature, unit output, other factors. This paper provides a scheme multi-unit collaborative optimization cooling. It mainly divided into four stages: first, built through Modelica, it exported as fmu file. Second, PSO strategy used to generate sets contain output cooling effect at different times. Third, LSTM TCN load forecasting models are trained. Finally, comparison various algorithms, PSO-LSTM-TCN simulate optimize parameters multiple units.