Large Language Model based Agent Framework for Electric Vehicle Charging Behavior Simulation
Junkang Feng, Chenggang Cui, Chuanlin Zhang, Zizhu Fan
TL;DR
This work addresses the challenge of accurately simulating EV charging behavior and its grid impact by introducing a generative, LLM-based agent framework. It integrates user preferences, psychological traits, and environmental factors through a seven-component design (L, C, P, M, S, A, R) and a memory-reflective loop, operating across scenario, time, space, energy, and price dimensions. The approach is demonstrated via a Shanghai taxi-driver setting, where 10 agents generate personalized profiles, dynamic daily plans, and adaptive charging decisions with continuous reflection; results are visualized on urban routes and charging-station choices. The framework offers practical implications for urban charging management by enabling real-time, personalized, and interpretable decision-making, with future work targeting more complex scenarios and broader data sources to enhance predictive accuracy.
Abstract
This paper introduces a new LLM based agent framework for simulating electric vehicle (EV) charging behavior, integrating user preferences, psychological characteristics, and environmental factors to optimize the charging process. The framework comprises several modules, enabling sophisticated, adaptive simulations. Dynamic decision making is supported by continuous reflection and memory updates, ensuring alignment with user expectations and enhanced efficiency. The framework's ability to generate personalized user profiles and real-time decisions offers significant advancements for urban EV charging management. Future work could focus on incorporating more intricate scenarios and expanding data sources to enhance predictive accuracy and practical utility.