An Efficient Hybrid Red Panda Optimization-Similarity-Navigated Graph Neural Network with Blockchain Technology for Shared Electric Vehicle Charging

Blockchain technology for shared electric vehicle (EV) charging makes use of a distributed ledger in handling different transactions which promotes authenticity in charging. However, this technology can be non-scalable at times which slows down the transaction rates as the scale increases in an organization and may also demand high computational as well as energy requirements which increases the inefficiency of the total system, the operating expenses of EV charging station, costs of energy utilized for charging, cost of maintenance and cost of station infrastructure may prove to be a weakness, impacting the feasibility of infrastructure deployment and upkeep expenses of the charging stations. To overcome these drawbacks, this paper proposes a hybrid method for blockchain technology with shared EV charging. The proposed approach combines the use of the Red Panda Optimization (RPO) with Similarity-Navigated Graph Neural Network (SNGNN). Hence, it is named as hybrid RPO-SNGNN approach. Reducing the operating cost of EV charging position and maximizes the security efficiency is the main aim of the proposed approach. To optimize scheduling and resource allocation by finding efficient paths by RPO and predict future demand and user behaviour patterns by SNGNN. By then, the proposed model has been put into practice on the MATLAB work platform and contrasted with other existing methods. The proposed technique outperforms all existing techniques, including Fuzzy-based Cross Hierarchical Reinforcement Q Learning (FCHRQL), Non-Dominated Sorting Genetic Algorithm-Long Short-Term Memory Network (NSGA-II-LSTMN), and Ant Colony Optimization-Proximal Policy Optimization (ACO-PPO). The proposed method shows the operating cost of $550 and the security rate is 97%. From the result, the proposed approach operating cost is low and the security rate is high when compared to the existing approaches.