Two-Stage Deep Reinforcement Learning for Inverter-Based Volt-VAR Control in Active Distribution Networks

Model-based Vol/VAR optimization method is widely used to eliminate voltage violations and reduce network losses. However, the parameters of active distribution networks(ADNs) are not onsite identified, so significant errors may be involved in model make model-based infeasible. To cope with this critical issue, we propose a novel two-stage deep reinforcement learning (DRL) improve profile by regulating inverter-based energy resources, which consists offline stage online stage. In stage, highly efficient adversarial algorithm developed train an agent robust mismatch. sequential transfer safely as perform continuous controlling significantly improved safety efficiency. Numerical simulations on IEEE test networks only demonstrate that proposed outperforms state-of-art algorithm, but also show our achieves much better performance than existing DRL based methods application.