Operation Optimization of Wind-Energy and Storage Combined Systems: A Study on Synergizing Multifaceted Application Values of Energy Storage

Energy Storage Systems (ESSs) are getting ever-increasingly employed in power systems because of their multifaceted application values, such as mitigating the negative impacts of and enhancing the accommodation capability for intermittent renewable energy generation. In this paper, the challenges posed by the inherent variability of Renewable Energy Sources (RESs) based generation are addressed by presenting an operation optimization model for Wind-energy and Storage Combined Systems (WSCSs). A Deep Reinforcement Learning (DRL) framework, specifically the Proximal Policy Optimization (PPO) algorithm and Gated Recurrent Unit (GRU) network, is employed to handle the dynamics and complexities of the electricity market. The proposed GRU-based PPO algorithm optimizes the scheduling of Battery Energy Storage Systems (BESSs) with an objective of maximizing their multifaceted application values, such as peak-valley arbitrage in an electricity market environment, error compensation, and fluctuation mitigation of intermittent renewable energy based power outputs. Finally, sample simulation results demonstrate the feasibility and efficiency of the proposed method in exploring multifaceted application values of ESSs.