Support vector machine based predictive controller with swarm intelligence for PEMFC

The modelling and control of the proton exchange membrane fuel cell (PEMFC) possess great challenges due to PEMFC´s inherent nonlinearities and time-varying properties. The objective of this paper is to propose a nonlinear model predictive control (MPC) strategy based on the support vector machine (SVM) and the particle swarm optimization (PSO). SVM is employed to establish the predictive model by mapping PEMFC performance as a function of operating conditions. PSO is then used to solve the optimization problem formulated by MPC. The SVM model and MPC strategy are implemented in the MATALB environment. Simulation results demonstrate the proposed control strategy can achieve robust control of PEMFC voltage with good performance in tracking reference trajectory.