Physics-informed Evolutionary Strategy based Control for Mitigating Delayed Voltage Recovery

TL;DR

This paper introduces a physics-informed, data-driven control method using evolutionary strategies to effectively mitigate delayed voltage recovery in power systems, improving robustness and training efficiency over traditional reinforcement learning approaches.

Contribution

The paper presents a novel physics-informed evolutionary strategy with a trainable action mask that enhances sample efficiency, robustness, and leverages physical knowledge for power system voltage control.

Findings

Outperforms existing methods in IEEE 300-bus system simulations.

Improves control robustness against unseen scenarios.

Reduces training time through surrogate gradient guidance.

Abstract

In this work we propose a novel data-driven, real-time power system voltage control method based on the physics-informed guided meta evolutionary strategy (ES). The main objective is to quickly provide an adaptive control strategy to mitigate the fault-induced delayed voltage recovery (FIDVR) problem. Reinforcement learning methods have been developed for the same or similar challenging control problems, but they suffer from training inefficiency and lack of robustness for "corner or unseen" scenarios. On the other hand, extensive physical knowledge has been developed in power systems but little has been leveraged in learning-based approaches. To address these challenges, we introduce the trainable action mask technique for flexibly embedding physical knowledge into RL models to rule out unnecessary or unfavorable actions, and achieve notable improvements in sample efficiency, control performance and robustness. Furthermore, our method leverages past learning experience to derive surrogate gradient to guide and accelerate the exploration process in training. Case studies on the IEEE 300-bus system and comparisons with other state-of-the-art benchmark methods demonstrate effectiveness and advantages of our method.