Explainable Home Energy Management Systems based on Reinforcement Learning using Differentiable Decision Trees

With the ongoing energy transition, demand-side flexibility has become important to provide grid support and allow further integration of sustainable energy sources.Residential energy assets constitute a major and largely untapped source of flexibility, driven by the increased adoption of solar PV, home batteries, and EVs.However, unlocking this residential flexibility is challenging, as it requires a control framework that can effectively manage household energy consumption while maintaining user comfort, which should be easily scalable across different, diverse houses.We aim to address this challenging problem and introduce a reinforcement learning-based approach using differentiable decision trees.Our proposed approach integrates the scalability of data-driven reinforcement learning with the explainability of (differentiable) decision trees.The resulting controller can be easily adapted across different houses and provides a simple control policy that can be explained to end-users, facilitating maximal user acceptance.As a proof-of-concept, we analyze our method using a home energy management problem, comparing its performance with commercially available rule-based baseline and conventional state-of-the-art neural network-based RL controllers.Our preliminary study indicates that the proposed method performs comparable to standard RLbased controllers, outperforming baseline controllers by ∼20% in terms of daily cost savings while being straightforward to explain. CCS CONCEPTS• Theory of computation → Reinforcement learning; •