H-FLTN: A Privacy-Preserving Hierarchical Framework for Electric Vehicle Spatio-Temporal Charge Prediction

2025-02-25

AI Summary

This work addresses the spatiotemporal forecasting of electric vehicle (EV) charging demand and mobile-driven network resource management. To balance privacy preservation and computational efficiency, we propose a three-tier federated learning framework. Methodologically, we innovatively integrate Dynamic Client Capacity Management (DCCM) with Client Rotation Management (CRM) to construct a hierarchical Federated Transformer model, reducing training time complexity from O(N) to O(1). Furthermore, we design a peer-to-peer (P2P)-enhanced additive secret sharing mechanism to ensure end-to-end privacy protection. Evaluated on large-scale real-world vehicular mobility datasets, our approach improves EV charging demand forecasting accuracy by 18.7%, significantly enhancing grid demand response capability and urban energy system stability.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Application Domains: Mobility, Driving & Flight

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Security and Privacy: Privacy-enhancing technologies

Abstract

The widespread adoption of Electric Vehicles (EVs) poses critical challenges for energy providers, particularly in predicting charging time (temporal prediction), ensuring user privacy, and managing resources efficiently in mobility-driven networks. This paper introduces the Hierarchical Federated Learning Transformer Network (H-FLTN) framework to address these challenges. H-FLTN employs a three-tier hierarchical architecture comprising EVs, community Distributed Energy Resource Management Systems (DERMS), and the Energy Provider Data Centre (EPDC) to enable accurate spatio-temporal predictions of EV charging needs while preserving privacy. Temporal prediction is enhanced using Transformer-based learning, capturing complex dependencies in charging behavior. Privacy is ensured through Secure Aggregation, Additive Secret Sharing, and Peer-to-Peer (P2P) Sharing with Augmentation, which allow only secret shares of model weights to be exchanged while securing all transmissions. To improve training efficiency and resource management, H-FLTN integrates Dynamic Client Capping Mechanism (DCCM) and Client Rotation Management (CRM), ensuring that training remains both computationally and temporally efficient as the number of participating EVs increases. DCCM optimises client participation by limiting excessive computational loads, while CRM balances training contributions across epochs, preventing imbalanced participation. Our simulation results based on large-scale empirical vehicle mobility data reveal that DCCM and CRM reduce the training time complexity with increasing EVs from linear to constant. Its integration into real-world smart city infrastructure enhances energy demand forecasting, resource allocation, and grid stability, ensuring reliability and sustainability in future mobility ecosystems.

Problem

• Predicts EV charging time and location accurately.
• Ensures user privacy during data sharing.
• Manages resources efficiently in EV networks.

Innovation

• Hierarchical Federated Learning
• Transformer-based temporal prediction
• Secure Aggregation privacy preservation