AI helps identify causes of fuel cell malfunctions: Study

A team of researchers has developed a novel way to analyse the microstructure of carbon fibre paper, a key material in hydrogen fuel cells, at a speed 100 times faster than existing methods, thanks to digital twin technology and artificial intelligence (AI).

Carbon fibre paper is a key material in hydrogen fuel cell stacks, playing a crucial role in facilitating water discharge and fuel supply. It is composed of materials such as carbon fibres, binders (adhesives) and coatings.

Dr Chi-Young Jung's research team from the Hydrogen Research and Demonstration Center at the Korea Institute of Energy Research (KIER) developed a technology that analyses the microstructure of carbon fibre paper using X-ray diagnostics and an AI-based image learning model.

Notably, this technology enables precise analysis using only X-ray tomography, eliminating the need for an electron microscope. As a result, it allows for near real-time condition diagnosis, according to the study published in journal Applied Energy.

The research team extracted 5,000 images from over 200 samples of carbon fibre paper and trained a machine learning algorithm with this data.

As a result, the trained model was able to predict the 3D distribution and arrangement of the key components of carbon fibre paper — including carbon fibers, binders, and coatings — with an accuracy of over 98 per cent.

“This study is significant in that it enhances analysis technology by combining AI with virtual space utilisation, and clearly identifies the relationship between the structure and properties of energy materials, thereby demonstrating its practical applicability,” said Dr Jung.

“We expect it to play a significant role in related fields such as secondary batteries and water electrolysis in the future,” he added.