Adaptive Feasibility Area Estimation to Enhance Cybersecurity of Electrolysis-Based Hydrogen Refueling Stations Integrated With Power Distribution Systems

This paper introduces a novel cyberattack-resilient model designed for the optimal operation management of electrolysis-based hydrogen refueling stations (eHRSs) integrated with electric power systems. The optimization model aims to coordinate the scheduling of eHRSs to concurrently support both the transportation sector and the electric power utility. This includes fulfilling the hydrogen demand of electric mobility systems (e-Mobility) and enhancing the resilience of the electric grid by following ancillary service signals issued by the grid operator. Adaptive feasibility areas (FAs) are estimated using the operating parameters of the integrated transportation and power system to identify potential cyberattacks. A framework is developed wherein dispersed eHRSs are managed by an eHRS chain aggregator. The operating parameters of eHRSs are communicated between the individual stations and the eHRS chain aggregator. Additionally, the eHRS chain aggregator interfaces with the electric power utility operator to address the utility’s requirements. Various scenarios are modeled to assess the technical and financial impacts of cyberattacks on eHRS. The proposed model is employed to detect false data injection attacks and mitigate the adverse effects of cyberattacks on the integrated transportation and power system. Simulation studies are conducted to evaluate the effectiveness and practicality of the proposed model. The performance of the FA-based method is compared with traditional deep neural network models and data-driven methods, demonstrating 13.5% and 8.68% improvements, respectively in detection accuracy. In addition, the proposed model achieves a 19% reduction in training time.