Recently, the Chengdu Branch of the National Engineering Research Center for Sensors, under the leadership of Professor Gao Bin from the School of Automation Engineering at the University of Electronic Science and Technology of China, released a research paper in IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS. This journal ranks in the first quartile according to the Chinese Academy of Sciences and is a top-tier publication in the field. The paper, titled "Physical and Digital Dual-Driven AI Framework for Enhanced Electromagnetic Perception in Nondestructive Testing Tomography," was authored with Chen Rui as the first author and Professor Gao Bin serving as the corresponding author.

The research introduces an innovative framework for electromagnetic nondestructive testing, which ingeniously blends physical and digital artificial intelligence elements. By integrating a physical electromagnetic neural network with a physically perceptive reinforcement learning engine, this framework facilitates real-time adaptive optimization of electromagnetic field sensing parameters. As a result, it significantly boosts the sensitivity and imaging resolution of defect detection.

Experimental findings reveal that the framework can achieve a maximum sensitivity improvement of 105.8% and is capable of quantifying defects as small as 0.2 millimeters. This performance outstrips that of existing electromagnetic nondestructive testing technologies. Moreover, the framework shows promising application prospects in fields involving electromagnetic sensing and control, such as controlled nuclear fusion and functional magnetic resonance imaging.