Multi-Objective Power Adjustment for Coupling Transmission Section Based on Deep Reinforcement Learning

As the scale of the power grid increases and the penetration of renewable energy resources increases, faster security assessment and control methods to maintain the active power of the transmission section within a reasonable range are required for large power grids. A power adjustment method based on multi-agent deep reinforcement learning is proposed in this paper, where each transmission section is controlled by one agent, and the multi-agent deep deterministic policy gradient (MADDPG) algorithm is adopted to obtain the control strategies. The rewards involve the renewable energy consumption, generation cost and the power flow on the transmission section within the limit. Due to the coupling relationship between transmission sections, auxiliary rewards are designed to improve the stability of agent training. The simulation verification is carried out based on the IEEE 39-bus system and a real power system. The method proposed in this paper can quickly give a section control strategy according to the current operating conditions, ensuring the safe and stable operation of the power grid.