Chung-Ang University Develops Technology to Restore Structural Defects Using Generative AI

Innovation in Restoration Expected for Architecture and Civil Engineering

Chung-Ang University announced on the 21st that the research team led by Professor Lee Sooyoung of the Department of Mechanical Engineering has developed a technology, in collaboration with the Korea Research Institute of Standards and Science (KRISS), that can restore and predict hidden defects inside structures using artificial intelligence (AI).

Image of generative AI-based technology for restoring hidden cracks in structures. Chung-Ang University

Technology capable of precisely monitoring internal defects in structures has been consistently demanded in various industries, including semiconductors, aviation, automotive, energy, and construction, to ensure safety and reliability.

While such technologies utilize physical sensors to detect damage or cracks inside structures, there have been limitations in fully identifying internal defects due to uncertainties such as resolution limits and signal distortion.

In response, the research team developed, for the first time in the world, a technology based on generative AI that can precisely restore hidden cracks inside structures in real time.

This technology represents a new approach, enabling generative AI to learn and overcome physical limitations on its own, virtually restoring and predicting the "invisible interior" that could not be observed with conventional sensors.

Furthermore, it can be directly applied as a real-time smart safety diagnostic technology in industrial settings that require high reliability, such as power plants, semiconductor equipment, construction structures, and aircraft.

Park Choonsoo, head of the Nondestructive Measurement Group at KRISS, stated, "This research is significant in that it enabled the precise restoration of complex internal defects through generative AI."

This research was supported by the National Research Foundation of Korea's Excellent Young Researcher Program and the Korea Research Institute of Standards and Science, and was published in the world-renowned mechanical engineering journal "Mechanical Systems and Signal Processing."

This content was produced with the assistance of AI translation services.