Evolving LLM-Derived Control Policies for Residential EV Charging and Vehicle-to-Grid Energy Optimization

2026-02-06

AI Summary

This work addresses the lack of interpretability in traditional reinforcement learning approaches to vehicle-to-grid (V2G) optimization, which produce black-box policies that fail to meet user and regulatory demands for transparent control. To overcome this limitation, the authors propose a novel method that integrates evolutionary computation with large language models (LLMs), embedding the LLM as an intelligent mutation operator within an evolutionary framework. Through a prompt-evaluate-repair loop, the approach automatically generates executable and human-readable Python charging strategies in the high-fidelity EV2Gym simulation environment. Without explicit programming, the method discovers sophisticated behaviors such as forward-looking arbitrage and hysteresis. Experimental results demonstrate that strategies generated using a hybrid prompting scheme achieve 118% of the baseline revenue while maintaining concise, auditable code suitable for real-world residential deployment.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Learning to Search
• Machine Learning: Evolutionary Learning

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

This research presents a novel application of Evolutionary Computation to the domain of residential electric vehicle (EV) energy management. While reinforcement learning (RL) achieves high performance in vehicle-to-grid (V2G) optimization, it typically produces opaque"black-box"neural networks that are difficult for consumers and regulators to audit. Addressing this interpretability gap, we propose a program search framework that leverages Large Language Models (LLMs) as intelligent mutation operators within an iterative prompt-evaluation-repair loop. Utilizing the high-fidelity EV2Gym simulation environment as a fitness function, the system undergoes successive refinement cycles to synthesize executable Python policies that balance profit maximization, user comfort, and physical safety constraints. We benchmark four prompting strategies: Imitation, Reasoning, Hybrid and Runtime, evaluating their ability to discover adaptive control logic. Results demonstrate that the Hybrid strategy produces concise, human-readable heuristics that achieve 118% of the baseline profit, effectively discovering complex behaviors like anticipatory arbitrage and hysteresis without explicit programming. This work establishes LLM-driven Evolutionary Computation as a practical approach for generating EV charging control policies that are transparent, inspectable, and suitable for real residential deployment.

Problem

• interpretability
• vehicle-to-grid
• electric vehicle charging
• control policy
• black-box models

Innovation

• LLM-driven Evolutionary Computation
• Interpretable Control Policies
• Vehicle-to-Grid Optimization
• Prompt Engineering
• EV2Gym Simulation