Authors:
Chibuzo V. Ikwuagwu

Addresses:
Department of Mechanical Engineering, University of Nigeria, Nsukka, Enugu, Nigeria. Centre of Excellence for Sustainable Sower and Energy Development, University of Nigeria, Nsukka, Enugu, Nigeria. 

Abstract:

Green hydrogen, produced through water electrolysis, has emerged as a key enabler of decarbonised energy systems, offering a clean and sustainable alternative to fossil fuels. The efficiency of electrolysers, however, is strongly dependent on multiple interacting process parameters, including temperature, pressure, and current density, which vary dynamically with operating conditions and energy input. Traditional approaches to optimising these parameters are often empirical, time-consuming, and may fail to identify globally optimal operating points. In this study, researchers present an AI-based multi-objective optimisation framework that enhances electrolyser efficiency by combining machine learning with evolutionary algorithms. A hybrid approach integrating experimental data with physics-informed surrogate models is used to train predictive models, which capture the nonlinear dependencies of system performance on operating variables. These models then guide the optimisation process, identifying operating regimes that improve efficiency by up to 15% relative to conventional setpoints. The results demonstrate the potential for cost-effective, scalable green hydrogen production, highlighting the value of AI-driven strategies for intelligent electrolyser operation and the sustainable energy transition.

Keywords: Green Hydrogen; Electrolyser Efficiency; Machine Learning (ML); Water Electrolysis; Fossil Fuels; Energy Input; Current Density; Sustainable Energy Transition.

Received on: 04/02/2025, Revised on: 09/04/2025, Accepted on: 06/07/2025, Published on: 03/03/2026

DOI: 10.69888/FTSESS.2026.000690

FMDB Transactions on Sustainable Environmental Sciences, 2026 Vol. 3 No. 1, Pages: 12-19

• Views : 38
• Downloads : 9

Download PDF