Performance Modeling of a Proton Exchange Membrane Hydrogen Fuel Cell Using an Artificial Neural Network : A Deep Learning-Based Approach

Abstract

Nowadays, hydrogen fuel cells occupy a crucial position in sustainable energy systems. However, a precise model of their performance is needed to improve their efficiency and integrate them into hydrogen electric vehicles. This paper presents a hydrogen fuel cell model based on artificial neural networks (ANN) to predict its performance characteristics. Using experimental data from a proton exchange membrane fuel cell (PEMFC) NEXA 1200 fuel cell in ISA laboratory, a feedforward ANN model optimized by deep learning has been developed, integrating advanced training techniques. The model's performance was evaluated on independent test sets, revealing predictive precision, with reduced mean-square error (MSE). A model development and evaluation will be reviewed in order to visualize the training progress and the results at the end of the simulation. The main advantages of the proposed feedforward ANN model lie in both its flexible architecture, which can capture complex relationships without the need for explicit physical models, and its predictive and optimization capability.