EV-STLLM: Electric vehicle charging forecasting based on spatio-temporal large language models with multi-frequency and multi-scale information fusion

2025-07-13

AI Summary

To address the challenges of complex spatiotemporal dependencies and insufficient representational capacity in electric vehicle (EV) charging demand and charging station occupancy forecasting, this paper proposes EV-STLLM—the first spatiotemporal large language model tailored for electric transportation. Methodologically, it innovatively integrates variational mode decomposition (VMD) with an improved ICEEMDAN algorithm for multi-frequency signal decomposition, employs fuzzy information granulation to extract multiscale features, and introduces a domain-knowledge-guided partially frozen graph attention mechanism. Furthermore, ReliefF-based feature selection and spatiotemporal frequency embedding are incorporated to enhance graph-structured representation learning. Extensive experiments on real-world data from Shenzhen demonstrate that EV-STLLM significantly outperforms state-of-the-art methods in both charging demand and station occupancy prediction, achieving up to a 12.6% improvement in accuracy while exhibiting superior robustness.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

With the proliferation of electric vehicles (EVs), accurate charging demand and station occupancy forecasting are critical for optimizing urban energy and the profit of EVs aggregator. Existing approaches in this field usually struggle to capture the complex spatio-temporal dependencies in EV charging behaviors, and their limited model parameters hinder their ability to learn complex data distribution representations from large datasets. To this end, we propose a novel EV spatio-temporal large language model (EV-STLLM) for accurate prediction. Our proposed framework is divided into two modules. In the data processing module, we utilize variational mode decomposition (VMD) for data denoising, and improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) for data multi-frequency decomposition. Fuzzy information granulation (FIG) for extracting multi-scale information. Additionally, ReliefF is used for feature selection to mitigate redundancy. In the forecasting module, the EV-STLLM is used to directly achieve EV charging and occupancy forecasting. Firstly, we fully capture the intrinsic spatio-temporal characteristics of the data by integrating adjacency matrices derived from the regional stations network and spatio-temporal-frequency embedding information. Then, the partially frozen graph attention (PFGA) module is utilized to maintain the sequential feature modeling capabilities of the pre-trained large model while incorporating EV domain knowledge. Extensive experiments using real-world data from Shenzhen, China, demonstrate that our proposed framework can achieve superior accuracy and robustness compared to the state-of-the-art benchmarks.

Problem

• Accurate EV charging demand and station occupancy forecasting
• Capturing complex spatio-temporal dependencies in EV charging behaviors
• Enhancing model ability to learn from large datasets

Innovation

• Uses VMD and ICEEMDAN for data denoising
• Integrates adjacency matrices for spatio-temporal features
• Employs PFGA to maintain pre-trained model capabilities