Predictive Nonlinear PID Neural Voltage-Tracking Controller Design for Fuel Cell based on Optimization Algorithm

Abstract

This paper proposes a predictive nonlinear PID neural voltage-tracking controller design for Proton Exchange Membrane Fuel Cell (PEMFC) Model with an on-line auto-tuning intelligent algorithm. The purpose of the proposed robust feedback nonlinear PID neural predictive voltage controller is to find the optimal value of the hydrogen partial pressure action in order to control the stack terminal voltage of the (PEMFC) model for one-step-ahead prediction. The Chaotic Particle Swarm Optimization (CPSO) is utilized as a stable and intelligent robust on-line auto-tuning algorithm to obtain the near-optimal weights for the proposed controller so as to improve the performance index of the system as well as to minimize the energy consumption. The Simulation results demonstrated the effectiveness of the proposed controller compared with the linear PID neural controller.