Energy management optimization strategy of virtual power plant based on deep reinforcement learning

Abstract The large-scale integration of new energy into the grid in future will have an impact on dispatching operation power grid. As a potential controllable resource, load is gradually being tapped, and application virtual plants artificial intelligence provides solution. It can effectively perform resource aggregation scheduling management optimization. Therefore, this paper proposes optimization strategy for based deep reinforcement learning. first establishes three types models storage then combines learning Double-DQN algorithm with internal model plant to construct environment, action, reward functions, finally, conducts simulation result analysis. calculation example shows that realize optimal improve demand side response, experimental has higher yield curve compared other algorithms, which verifies effectiveness rationality strategy. significant inspire promote high-quality development green economy ideas models.