Research on Day-ahead Optimization Operation Strategies for Data Centers in Scenarios of Orderly Electricity Use

With the increasing energy consumption of data centers, its impact on the power system is becoming more and more significant. Data centers are not only energy-intensive loads, but also important waste heat producers, with supply and demand bidirectional. At present, the operation of data center micro-energy network is complicated, and there are problems such as the difficulty of fine load modeling of data center and the difficulty of integrating demand response resources to participate in the power market. It makes it difficult for data centers to provide reliable regulation capabilities, resulting in orderly power consumption difficulties. In this paper, a data center load model is established to calculate the time-space transfer characteristics of the load in the day-ahead market, including the IT equipment energy consumption model and the cooling system energy consumption model, as well as the calculation load and server operation model. Then, based on the power market price signal, the optimal scheduling model of multi-location data center is established to minimize the total cost of multi-location data center, and the optimal distribution of computing load is realized.