Abstract

This paper presents two systematic machine learning (ML) approaches for predicting the hydrogen production rate and cell current density for proton exchange membrane electrolyzer cells (PEMEC) as a function of different design parameters. Management of the hydrogen production rate and current density remains an important research topic for many technologies, including PEM water electrolysis cells, which are the focus of this study; therefore, two applications of ML were proposed to simulate the hydrogen production rate and current density for PEMEC. In this study, five different ML models were trained and tested for each simulation i.e., artificial neural networks, polynomial regression, support vector machine regressor, K-nearest neighbor regressor, and decision tree regressor, using 15 different input parameters and single output parameters. Box and whisker data analysis was applied to obtain the highest activation materials for the current density in the dataset. The box whisker plots revealed that Nafion115 and Nafion117 for the membrane type, titanium for the electrode type (cathode and anode), platinum (Pt) as the cathode catalyst, Pt and ruthenium as anode catalysts, and deionized water and methanol as the electrolyte (catholyte and anolyte) were the highest activation materials for current density. The performance comparison of the ML models was given by calculating the mean absolute error for the training and test data for each model. For both simulations, the ANN model showed the best performance, with a mean absolute error of 5.0006 for training and 6.4383 for testing for hydrogen production simulation, and 0.03819 for training and 0.04036 for testing for the current density simulation. Both applications have minimal error values, which means that the proposed methods simulate the real PEM cell performance well. This study paves the way for a significant method to reduce the cost, effort, and time required to fabricate PEMEC.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call