Predicting PEMFC maximum power point using an artificial neural network

This paper presents an artificial neural network (ANN) to predict the maximum power point (MPP) of a proton exchange membrane fuel cell (PEMFC). First, the fuel cell is modeled and simulated using the Amphlett model. Then, we develop a tracking algorithm for maximum power point based on a Radial Basis Function (RBF) Artificial Neural Network (ANN) in order to calculate the maximum power point defined by its current and voltage whatever the fuel cell operating conditions (oxygen and hydrogen partial pressures, operating temperature and membrane hydration rate). The Amphlet PEMFC electrochemical model is used to generate the training examples of the proposed neural network and to validate the obtained results. The results show that the ANN can accurately predict the fuel cell current and voltage at the point of maximum power for all operating conditions.