Sustainable Electric Vehicle Design with Integrated Fuel Cell, Photovoltaic, and ANN-Based Model Predictive Control

Abstract

  This paper presents a new Electric Vehicle (EV) system that combines a fuel cell, electrolyzer, onboard PV cell, and a quadratic bidirectional buck-boost converter. The system is managed by Model Predictive Control (MPC) with support from an Artificial Neural Network (ANN). The PV cell enhances the fuel cell's power output during operation and, during idle periods, powers the electrolyzer to produce hydrogen for storage. This stored hydrogen is used to fuel the vehicle, improving its energy efficiency and reducing reliance on traditional fuel sources. A quadratic bidirectional buck-boost converter is used to control voltage, improving power management between the motor and energy sources. This ensures efficient execution even when fluctuating irradiance and vehicle speeds vary. An enhanced Maximum Power Point Tracking (MPPT) algorithm, utilizing an incremental conductance method, is used to maximize energy extraction from the PV system. Meanwhile, the converter’s dual-loop control, featuring outer voltage and inner current control, ensures a stable DC output for motor operation. The vehicle drive system directs an indirect vector-controlled induction motor, which is set by the ANN-based MPC. This system employs a novel predictive torque control strategy that eliminates the need for weighting factors, simplifying the control process while improving motor speed regulation. Additionally, an ANN s used to accurately predict motor speed based on real-time stator current data, ensuring optimal performance and system stability. Simulations conducted in MATLAB/SIMULINK evaluate the system's performance under varying solar irradiance and speed conditions. The results determine the feasible of the proposed EV configuration to enhance energy efficiency, improve power quality, and provide a more sustainable and economically viable solution for electric vehicles by integrating renewable energy sources with advanced control techniques.