Partially Observable Residual Reinforcement Learning for PV-Inverter-Based Voltage Control in Distribution Grids

TL;DR

This paper presents a novel Residual Reinforcement Learning framework that improves voltage control in distribution grids by enabling faster learning and practical deployment using inverter measurements alone.

Contribution

It introduces a residual RL approach with a specialized neural network architecture that enhances convergence speed and practicality for real-world voltage regulation.

Findings

Faster convergence compared to traditional RL methods

Effective voltage regulation with minimal power curtailment

Operates solely on inverter measurements without full grid state info

Abstract

This paper introduces an efficient Residual Reinforcement Learning (RRL) framework for voltage control in active distribution grids. Voltage control remains a critical challenge in distribution grids, where conventional Reinforcement Learning (RL) methods often suffer from slow training convergence and inefficient exploration. To overcome these challenges, the proposed RRL approach learns a residual policy on top of a modified Sequential Droop Control (SDC) mechanism, ensuring faster convergence. Additionally, the framework introduces a Local Shared Linear (LSL) architecture for the Q-network and a Transformer-Encoder actor network, which collectively enhance overall performance. Unlike several existing approaches, the proposed method relies solely on inverters' measurements without requiring full state information of the power grid, rendering it more practical for real-world deployment. Simulation results validate the effectiveness of the RRL framework in achieving rapid convergence, minimizing active power curtailment, and ensuring reliable voltage regulation.