Context-aware energy auctions on blockchain for mobile EV charging

The rapid growth of EVs on the road has outpaced the expansion of charging infrastructure, highlighting the need for innovative solutions like Mobile Charging Stations (MCS). These on-the-go chargers are vital for regions with minimal infrastructure but require careful planning to maximize throughput and profit while ensuring customer satisfaction. One key challenge is customer no-shows, leading to resource wastage in terms of forced idle time at MCS charging points. This paper proposes a machine learning-based, context-driven EV behavior prediction model that optimizes MCS decision-making during energy auctions, where MCS act as bidders. By integrating environmental factors like weather, traffic, and time with driver behavior history, the system predicts EV charging patterns, improving resource utilization. The proposed solution is presented as a zone-distributed framework that is deployed as a blockchain-based Decentralized Application (DApp). Experimental results demonstrate the approach's efficiency in reducing wasted time and maximizing revenue for MCS operations.