Residual Deep Reinforcement Learning for Inverter-based Volt-Var Control

TL;DR

This paper introduces a residual deep reinforcement learning method for inverter-based Volt-Var control in active distribution networks, combining model-based optimization with RL to improve control accuracy and learning efficiency.

Contribution

It proposes a novel RDRL framework that integrates model-based optimization with residual policy learning, and extends it to a boosting RDRL for enhanced performance.

Findings

RDRL inherits the control capability of approximate model-based optimization.

Residual policy learning enhances the policy optimization process.

Boosting RDRL further improves optimization performance with a smaller residual action space.

Abstract

A residual deep reinforcement learning (RDRL) approach is proposed by integrating DRL with model-based optimization for inverter-based volt-var control in active distribution networks when the accurate power flow model is unknown. RDRL learns a residual action with a reduced residual action space, based on the action of the model-based approach with an approximate model. RDRL inherits the control capability of the approximate-model-based optimization and enhances the policy optimization capability by residual policy learning. Additionally, it improves the approximation accuracy of the critic and reduces the search difficulties of the actor by reducing residual action space. To address the issues of "too small" or "too large" residual action space of RDRL and further improve the optimization performance, we extend RDRL to a boosting RDRL approach. It selects a much smaller residual action space and learns a residual policy by using the policy of RDRL as a base policy. Simulations demonstrate that RDRL and boosting RDRL improve the optimization performance considerably throughout the learning stage and verify their rationales point-by-point, including 1) inheriting the capability of the approximate model-based optimization, 2) residual policy learning, and 3) learning in a reduced action space.