Machine learning predicts voltage decline in water electrolyzers
An Efficient Machine Learning Approach for Degradation Forecasting in AEM Water Electrolysis
Machine Learning
Summary
Water electrolyzers split water into hydrogen and oxygen but lose efficiency over time. The authors used machine learning models to predict how the voltage needed to run these devices changes as they age. They tested several types of models on a new dataset from various experiments. This approach helps foresee equipment degradation, which can improve maintenance and reliability.
What this means in practice
- •For industrial maintenance teams: Predict voltage degradation in water electrolyzers to plan maintenance and avoid unexpected failures.
- •For energy systems developers: Incorporate predictive degradation models to enhance the design and operation of hydrogen production systems.
Authors
Marco Veneriano, Ani Gjergji, Sebastiano Bellani, Andrea Riva, Vito Paolo Pastore, Matteo Santacesaria
Abstract
This study provides a data-driven analysis of a novel dataset of single-cell Anion Exchange Membrane water electrolyzers (AEMWE), operated under constant current load across multiple heterogeneous experimental campaigns. We train and evaluate a range of machine learning models with different complexity, including linear baselines, LSTMs and CNNs, to perform medium-term forecasting of the cell voltage degradation curve. The models are assessed within a rigorous training and evaluation framework specifically designed for heterogeneous industrial data.