Adaptive maximum power extraction technique in fuel-cell integrated with novel DC-DC converter topology for low-power electric vehicle applications

Abstract

High gain single switch DC-DC converter for Proton Exchange Membrane Fuel Cell (PEMFC) with Neural-Network (NN) algorithm based Maximum Power Point Tracking (MPPT) technique to drive Electric Vehicle (EV) is proposed in this paper. Apart from many maximum power extraction techniques, Neural Network (NN) based algorithm control is proposed, which uses Radial Basis Function Network (RBFN) to harvest maximum power from PEMFC under different temperature variations, which is compared with traditional Incremental Conductance (InC) method. Apart from it, a novel Single Ended Primary Inductor (SEPI) converter (boost converter) with the Coupled Inductor and Switched-Capacitor Circuit (CI + SCC) is designed to lower stress in the voltage across the switch with improved voltage gain to meet drive train (Three-Phase BLDC Motor) scenarios. Proposed PEMFC performance with RBFN-based MPPT technique integrated to drive train using a novel high gain converter is simulated using MATLAB/Simulink and implemented using dSPACE-DS1104 real-time controller. The compact sizing of the converter is proposed to integrate with PEMFC, which would be suitable for low-power EV applications.

Keywords

• Fuel cell
• Radial Basis Function Neural Network
• Modified CI+LD technology
• Real-time controller
• SEPI Converter
• Brushless DC (BLDC) Motor