Abstract: With the widespread deployment of data centers, addressing their high energy consumption is becoming a major technical challenge. In order to reduce energy costs of data center and alleviate environmental pollution problems, using renewable energy to supply power for data centers is becoming the focus of attention. However, due to the randomness and volatility of the renewable energy, the consumption rate of renewable energy in data centers is relatively low. On the premise of considering the random fluctuation characteristics of the renewable energy and the load controllable characteristics of the data center cooling system, a variable temperature control strategy for the instant consumption of renewable energy in the data center cooling system is proposed. Through matching the cooling power consumption of the data center with the power generation of renewable energy, the cooling capacity changes with the output of renewable energy. The thermal inertia of the data center room can be used to attenuate and delay temperature changes to ensure that the data center maintains a safe operating temperature when the cooling capacity fluctuates. The data center energy consumption model and temperature dynamic response model are built, and a new energy efficiency index suitable for evaluating the utilization rate of renewable energy in the data center is proposed. It is verified that the proposed variable temperature control strategy can significantly improve the utilization rate of renewable energy and reduce the consumption of utility electricity.