New Neural Network Single Sensor Variable Step Size MPPT for PEM Fuel Cell Power System

Abstract

This paper deals with the development of a single sensor neural network controller used to track the maximum power of proton exchange membrane fuel cell power system. The proposed single sensor neural network controller has been developed and trained using single sensor maximum power point tracking data obtained previously. The developed maximum power point tracking controller has been used to track the output power of the fuel cell power system composed of 7 kW proton exchange membrane fuel cell powering a resistive load via a DC-DC boost converter controlled using the proposed controller. Simulation results obtained using the developed MATLAB/Simulink model show that the proposed single sensor neural network maximum power point tracking controller can track effectively the maximum power using only one sensor compared to the classical power point tracking controllers using two sensors reducing by the way the cost and the complexity of the fuel cell maximum power tracking controller.