Design and experimental validation of an artificial neural network-SVPWM controller for a novel micro grid-tied fuel cell-based 3-phase boost inverter

Abstract

A grid-tied fuel cell (FC) system demands efficient power conversion, power quality preservation, grid stability, power flow management, renewable energy source (RES) integration, and enhanced grid resilience. Achieving these goals requires a precise inverter circuit switch approach. A boost power converter connected to the FC stack ensures voltage regulation, power conditioning, efficient power transfer, system integration, control, and protection. This enhances FC system adaptability and compatibility across various applications, minimizing input current ripples for prolonged FC lifespan.This study introduces a novel DC-DC boost converter with an artificial neural network (ANN) controller to reduce FC input current ripples and enhance FC stack-generated voltage for grid applications. It also presents a space vector sinusoidal pulse width modulation (SVPWM) technique for FC-based three-phase grid-tied inverters. This offers improved voltage utilization, precise voltage and current control, reduced harmonic distortion, rapid response, flexibility, scalability, reduced total harmonic distortion (THD), and over-modulation capability. The proposed SVPWM technique utilizes a digital signal processing (DSP)-based controller, combining high-speed processing, precision, and real-time capabilities to enhance system performance and efficiency.