Key points are not available for this paper at this time.
Hydrogen energy is considered pivotal for achieving a carbon-neutral future, yet it faces significant challenges in terms of cost-effective production, efficient storage, and scalable catalyst development. Traditional experimental and computational approaches are often slow and resource-intensive. The integration of artificial intelligence (AI) and machine learning (ML) is rapidly transforming the hydrogen technology landscape by enabling accelerated discovery and optimization processes. This review comprehensively discusses the recent advances in AI for catalyst discovery in hydrogen evolution and oxygen evolution reactions and its role in optimizing hydrogen production systems, such as electrolysis, photocatalysis, and thermochemical water splitting. Additionally, it covers machine-learning-driven insights into hydrogen storage materials, including metal hydrides, metal-organic frameworks, and liquid organic hydrogen carriers. The key challenges, data strategies, and emerging techniques such as explainable AI, generative models, and digital twins are also addressed. The review concludes with future directions and perspectives on how AI and ML can revolutionize the hydrogen economy.
Sharma et al. (Wed,) studied this question.