# An Intelligent Condition-Monitoring Framework for Alkaline Water Electrolyzers Based on Hybrid Physics-Informed Health Indicators

## TL;DR

This paper introduces a smart monitoring system for water electrolyzers using physics-informed machine learning to improve safety and efficiency in green hydrogen production.

## Contribution

The novel hybrid framework combines physics-based models with machine learning to create Health Indicators for AWEs with limited sensor data.

## Key findings

- A CFD model was developed and validated to simulate AWE behavior under various conditions.
- An MLP model achieved 90.43% accuracy in real-time health state classification of AWEs.
- The framework enables predictive maintenance for AWEs under volatile renewable energy inputs.

## Abstract

What are the main findings?
A hybrid physics-informed machine learning (ML) framework is proposed for constructing Health Indicators (HIs) and enabling intelligent condition monitoring of Alkaline Water Electrolyzers (AWEs).Trained on a CFD-generated dataset, a Multilayer Perceptron (MLP) model achieves 90.43% accuracy in real-time health state classification, serving as an effective intelligent monitoring agent.

A hybrid physics-informed machine learning (ML) framework is proposed for constructing Health Indicators (HIs) and enabling intelligent condition monitoring of Alkaline Water Electrolyzers (AWEs).

Trained on a CFD-generated dataset, a Multilayer Perceptron (MLP) model achieves 90.43% accuracy in real-time health state classification, serving as an effective intelligent monitoring agent.

What are the implications of the main findings?
The proposed methodology provides a practical solution for predictive maintenance of AWEs operating under volatile renewable energy, enhancing system safety and reliability.It demonstrates the significant potential of combining mechanistic models with machine learning for intelligent monitoring in complex industrial systems where sensor data is limited.

The proposed methodology provides a practical solution for predictive maintenance of AWEs operating under volatile renewable energy, enhancing system safety and reliability.

It demonstrates the significant potential of combining mechanistic models with machine learning for intelligent monitoring in complex industrial systems where sensor data is limited.

Alkaline Water Electrolyzers (AWEs) are critical for green hydrogen production but face operational risks due to volatile renewable energy inputs. This study proposes an intelligent condition-monitoring framework that leverages a hybrid physics-informed machine learning (ML) methodology to construct Health Indicators (HIs). The core innovation lies in addressing the challenge of inaccessible internal states. First, a high-fidelity Computational Fluid Dynamics (CFD) model is developed and experimentally validated, serving as a physics-informed data generator to simulate multiphysics behavior under various operating and fault conditions. From this reliable simulation basis, a comprehensive dataset is produced, and eight key operational parameters are derived as HIs. This dataset is then used to train and benchmark three ML models for rapid health state classification. The Multilayer Perceptron (MLP) model achieves superior performance with 90.43% accuracy, effectively translating the validated physical understanding into a fast, deployable intelligent monitoring agent. This work presents a viable pathway for constructing reliable HIs and implementing AI-enhanced condition monitoring for AWEs, contributing to safer and more efficient green hydrogen production.