Work

Machine learning applications in semiconductor metal oxide chemiresistive hydrogen sensing: A review

Year: 2026

Type: article

Abstract: Hydrogen (H 2 ) is considered as a promising substitute energy carrier for fossil fuels. Such systems need reliable sensors for safety monitoring due to its high flammability and explosive properties. Typically, semiconductor chemiresistive sensors are widely employed in different applications. However, the demand for high-performance sensors based on such chemiresistive materials remains largely unmet. Recently, machine learning (ML) approaches have been broadly used to enhance the H 2 sensing performance of these chemiresistive sensors. This review article explores such approaches in detail. Despite these advancements, challenges remain in achieving consistent sensing performance, particularly under significant variations in environmental conditions. Future research directions will include improving the overall sensing performance of these chemiresistive sensors while improving existing sensing platforms in combination with advanced ML algorithms. Such efforts will pave the way for safer H 2 utilisation in a sustainable energy future. • Chemiresistive hydrogen gas sensing technologies and applications. • Critical review in machine learning for chemiresistive hydrogen gas sensing. • Limitations and future directions in hydrogen-specific chemiresistive sensing.

Source: International Journal of Hydrogen Energy

DOI: 10.1016/j.ijhydene.2026.155132

Authors: Sandun Ranasinghe, Ali Yavari, Thilini Thathsara, Christopher J. Harrison, Mahnaz Shafiei

Institutions: Swinburne University of Technology, RMIT University

Language: English

Topic: Gas Sensing Nanomaterials and Sensors

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,730

Funders: Swinburne University of Technology, Australian Renewable Energy Agency