Empowering Consumer Electric Vehicle Mobile Charging Services With Secure Profit Optimization

Abstract

The rapid expansion of Intelligent Transportation System (ITS) services depends on the Electric Vehicle (EV) and Mobile Charging Station (MCS) consumer electronics industry, as well as the intelligent Consumer Internet of Things (CIoT) platform. The functioning environment of MCSs is inherently dynamic, influenced by inconstant user preferences, energy demands, and charging service availability. Adapting to these changes in near-real-time while ensuring cost efficiency and fairness poses a notable challenge. These consumer electronic devices share data with third parties, so privacy is a critical concern. This paper presents a secure, optimized approach for enhancing the performance and accuracy of charging/discharging scheduling of MCSs within the CIoT network while protecting consumers’ data. This study aims to develop an optimization mechanism that enables decentralized learning with minimal data transfer while preserving user privacy by embedding Federated Learning (FL) as a security layer in our system. Also, it aims to maximize the potential profit of these stations while optimizing their daily operational efficiency. We propose a fog-edge communication to enhance communication in the decentralized FL-based network. Evaluating the result demonstrated enhanced profit maximization for MCSs operating within the CIoT network to fulfill as many energy requests from EVs as feasible while reducing self-charging expenses.