Robust and secure cloud-based monitoring of PV based electric vehicle system using novel optimized neural network

This research addresses the sustainable electrification of transportation through Photovoltaic (PV)-based Electric Vehicles (EVs), supported by a secure, cloud-based monitoring system. Real-time EV diagnostics, battery analytics, and PV system data are collected using IoT devices and transmitted securely via an Improved Diffie-Hellman (IDH) and Twofish Encryption (TE) technique to the cloud. A novel Crayfish Optimization Algorithm (COA)-based Deep Feed Forward Neural Network (DFFNN) identifies optimal routing paths, minimizing cost, delay, and energy consumption. For enhancing PV performance, a High Gain Quadratic Boost Converter (HGQBC) is implemented achieving 96.29 % efficiency with minimal input ripple, while an Improved Incremental Conductance (IC) Maximum Power Point Tracking (MPPT) ensures 99.61 % tracking efficiency with a 0.1 s response time. MATLAB simulations and laboratory prototypes validate the system’s efficacy, showcasing superior packet delivery ratios, low energy use, and fast response times, making this approach a robust solution for PV-based EV optimization.