Flexible Load Scheduling of Intelligent Buildings Based on Deep Reinforcement Learning

Intelligent buildings include scattered and large amount of flexible load, facing a large number of air conditioning equipment, the model mechanism is complex, and the traditional optimizer takes a long time to solve. In order to solve the above problems, this paper proposes the deep reinforcement learning method to optimize the running state of load. Firstly, the framework of the building load control system in the current popular stage is proposed, and the optimization process is introduced. Secondly, a reinforcement learning model applicable to the building is built, and the equipment status and real-time electricity price information are obtained in real time. The efficient processing capacity of the reinforcement learning algorithm is used to optimize the equipment status in the next period. Finally, a building is simulated and compared with the traditional optimizer to verify the effectiveness of the proposed method.