Energy System Optimization Research Based on PEC-DDPG Algorithm

Abstract Factors such as the stochastic nature of loads in energy systems make it difficult to optimize operation integrated systems. To address these problems, an system economy optimization scheme based on PEC-DDPG is proposed. Firstly, exponential moving average (EMA) introduced into deep deterministic policy gradient (DDPG) algorithm, and prioritized experience replay (PER) added pool prioritize improve learning efficiency overestimation existing a single Critic network solved by using multi-Critic structure. Next, model constructed, appropriate observation states, decision actions reward functions are selected. Finally, simulations data region show that better than operational DDPG algorithm.