A Model-Free Scheduling Method for Heterogeneous Electric Vehicle Cryptographic Task Processing Architecture

In recent years, Electric Vehicles (EVs) have become widely used due to their environmental friendliness. However, the increase in essential supporting Electric Vehicle Supply Equipments (EVSEs) has led to a large and diverse cryptographic demand for charging aggregators. The growing demand for computational power cannot be met by local cryptographic computing clusters, while cryptographic computing services on the cloud bring risks such as cross-VM (cross-Virtual-Machine) attacks. Meeting the requirements of security and timeliness simultaneously poses challenges for cryptographic service architecture and scheduling algorithms. To address these challenges, a heterogeneous task processing architecture is proposed to integrate local and cloud computing resources. Next, a task queue is constructed to transform the dynamic scheduling problem of multiple cryptographic tasks into a queue-jumping sequential decision-making problem. By modeling security requirements as constraints, the problem is modeled as a Constrained Markov Decision Process (CMDP) formulation. A Deep Reinforcement Learning (DRL) method with recurrent networks and a safe action mechanism are proposed to solve the formulation, which utilizes the temporal characteristics of cryptographic tasks to achieve better scheduling performance under constraints. Finally, compared with the state-of-the-art method, our proposed approach is validated through simulation experiments, demonstrating superior security and enhanced processing efficiency across various scenarios.