Dynamic model-based hardware in the loop simulation of PEMWE system for real-time fault diagnosis

The increasing deployment of proton exchange membrane water electrolysis (PEMWE) systems for renewable hydrogen production requires reliable real-time fault diagnosis under dynamically varying operating conditions. This study proposes a dynamic model-based hardware-in-the-loop simulation (HILS) framework for real-time fault diagnosis of a PEM water electrolysis system. A physics-based dynamic model of a 300 W class PEMWE system was developed and validated against experimental data under steady-state and transient conditions, demonstrating good agreement in voltage and temperature responses. Based on the validated model, representative system-level fault scenarios were systematically generated and used to develop a real-time fault diagnosis model capable of identifying single and multiple fault conditions. The proposed framework was implemented in a HILS environment using a PXI-based real-time platform and evaluated with live experimental data. The results confirm reliable detection and isolation of pump and sensor faults while accounting for dynamic thermal behavior and realistic timing constraints. The proposed HILS-based framework provides a practical solution for enhancing the reliability and safety of PEM water electrolysis systems in renewable hydrogen production applications. • A dynamic model of a PEM water electrolysis system was developed and validated. • System-level fault scenarios were generated using a model-based approach. • A real-time fault diagnosis model for single and multiple faults was proposed. • The diagnosis framework was implemented using hardware-in-the-loop simulation. • Experimental results confirmed reliable real-time fault detection performance.