부산대학교 도서관

Base on the virtual power plant (VPP), this paper studies the regulation strategy of using user-side energy storage as a backup power source to provide power supply for the park when the external power grid fails in the industrial park. Aiming at the continuity of the output of wind power plants, photovoltaic power plants and energy storage power sources, this paper uses an improved depth deterministic strategy gradient algorithm (DDPG). Based on the temporal characteristics of the external environment and the agent, the article uses a cyclic neural network to replace the original convolutional neural network, and at the same time uses a multi-simulator parallel processing strategy to accelerate processing. The calculation results show that, compared with the traditional DDPG algorithm, the strategy adjustment strategy described in this article has obtained a better adjustment effect.