Deep Reinforcement Learning-Based Robust Protection in DER-Rich Distribution Grids

This paper introduces a new framework of deep reinforcement learning based protective relay design in power distribution systems with many distributed energy resources (DERs). With increasing penetration electronically-interfaced resources, conventional overcurrent relays’ performance is rendered less effective due to the two-way uncertainties flow patterns. In this paper, machine learning-based that designed for adaptively deciding threshold action proposed. The particular algorithm used an Long Short-Term Memory (LSTM) enhanced neural network highly accurate, communication-free and easy implement. proposed tested OpenDSS simulation on IEEE 34-node test feeder collection large synthetic feeders Austin, Texas area. By designing adaptability upfront, shown substantially improve terms failure rate, robustness, response speed, scenarios high level resources.