Methodology For Optimal Parametrization Of The Polymer Membrane Fuel Cell Based On Elman Neural Network Method And Quantum Water Strider Algorithm

The present study suggests a new approach to optimal output voltage estimation of the proton exchange membrane fuel cell (PEMFC) based on a new optimized Elman Neural Network (ENN). Based on the proposed method, the ENN has been optimized based on a new version of the improved metaheuristic method, called Quantum Water Strider Algorithm (QWSA), and is utilized to define the main parameters of the ENN. The purpose is for minimizing the error value between the measured and the predicted output voltages of the PEMFC. To illustrate the advantage of the suggested model, its results on test equipment are compared with some other methods from the literature including genetic algorithm-based, Fluid search optimization-based, and the original WSA to indicate its effectiveness. The results showed that the proposed method with 0.0466 absolute error has the best agreement with the real value. The genetic algorithm-based, the Fluid search optimization algorithm-based, and the original WSA-based methods with 0.0574, 0.0658, and 0.0742 absolute error are in the next ranks (C) 2021 The Authors. Published by Elsevier Ltd.