Explainable AI: Deep Reinforcement Learning Agents for Residential Demand Side Cost Savings in Smart Grids

TL;DR

This paper presents an explainable deep reinforcement learning approach for managing household energy storage in smart grids, aiming to maximize cost savings while providing insights into the agent's learning process.

Contribution

It introduces a method to interpret the learning progression and strategies of a deep RL agent in residential energy management, addressing the black box nature of typical models.

Findings

The RL agent effectively learns to optimize energy storage usage.

The explanation of the agent's strategies enhances transparency.

The approach achieves demand-side cost savings in smart grids.

Abstract

Motivated by recent advancements in Deep Reinforcement Learning (RL), we have developed an RL agent to manage the operation of storage devices in a household and is designed to maximize demand-side cost savings. The proposed technique is data-driven, and the RL agent learns from scratch how to efficiently use the energy storage device given variable tariff structures. In most of the studies, the RL agent is considered as a black box, and how the agent has learned is often ignored. We explain the learning progression of the RL agent, and the strategies it follows based on the capacity of the storage device.