Remaining Useful Life Prediction of Proton Exchange Membrane Fuel Cells Using Genetic Algorithm Based Nonlinear Autoregressive Exogenous Network

Abstract

The proton exchange membrane fuel cell (PEMFC) is one of the most promising clean energy sources with high energy conversion, no electrolyte leakage and low operating temperature. However, it faces a limited lifetime due to degradation under normal operating conditions, and uncontrolled excessive degradation may even lead to catastrophic failures, such as explosions. Therefore, the importance of accurate estimation of the remaining useful life (RUL) cannot be overemphasized. A joint prediction method based on a genetic algorithm (GA) and a nonlinear autoregressive neural network (NARX) with external inputs is proposed. The proposed method is trained and validated with the 2014 IEEE PHM Data Challenge dataset, and compared with two common artificial neural network algorithms: genetic algorithm-based backpropagation neural network (GA-BPNN) and genetic algorithm-based time delay neural network (GA-TDNN). The results show that the proposed method has better prediction accuracy compared with the other two artificial neural network algorithms.