Robust Nonlinear Observer for State Estimation of Proton Exchange Membrane Fuel Cell Systems

Abstract

The proton exchange membrane fuel cell (PEMFC) system has strong nonlinearity and coupled characteristics. In order to better improve the control efficiency and dynamic performance of PEMFC system, a robust nonlinear observer for the oxygen excess ratio (OER) estimation is proposed in this paper. For PEMFC system, lots of system parameters can't be measured directly during the fuel cell operating, including air mass in the supply manifold, oxygen mass and nitrogen mass in the cathode. The proposed method can be applied to effectively solve the problem of estimating the state variables with a high accuracy. Through matrix transformations, the differential mean value theorem is used to transform the fuel cell system into a linear variable parameter system. Based on the Lyapunov stability theory and convexity principle, the robust observer design problem can be regarded as a linear matrix inequality(LMI) feasibility problem. The simulation results indicate that the designed observer can guarantee the convergence properties, and system control can be strengthened.