Seoul, Dec 31 (IANS) A team of researchers analyzes the microstructure of carbon fiber paper, a key material in hydrogen fuel cells, 100 times faster than existing methods thanks to digital twin technology and artificial intelligence. Developed a new method. Intelligence (AI).
Carbon fiber paper is the main material in hydrogen fuel cell stacks and plays an important role in facilitating water drainage and fuel supply. It consists of materials such as carbon fiber, binder (adhesive), and coating.
Dr. Chi-Young Jung’s research team at the Korea Institute of Energy Research (KIER) Hydrogen Research and Demonstration Center has developed a technique to analyze the microstructure of carbon fiber paper using X-ray diagnostics and an AI-based image learning model.
In particular, this technology does not require an electron microscope and enables precise analysis using only X-ray tomography. The result is near real-time condition diagnosis, according to research published in the journal Applied Energy.
The research team extracted 5,000 images from more than 200 carbon fiber paper samples and used this data to train a machine learning algorithm.
As a result, the trained model was able to predict the 3D distribution and placement of key components in carbon fiber paper, including carbon fibers, binder, and coating, with more than 98% accuracy.
Dr. Jung said, “This research is significant in that it combined AI and the use of virtual space to strengthen analysis technology, clearly identified the relationship between the structure and properties of energy materials, and demonstrated the possibility of practical application. ” he said.
“In the future, it is expected to play an important role in related fields such as secondary batteries and water electrolysis,” he added.
