Magnetic Barkhausen Noise (MBN, Magnetic Barkhausen Noise) technology, as a convenient and efficient non-destructive testing method, can detect and characterize the fatigue damage of ferromagnetic materials, but the random nature of the MBN signal makes its magnetic characteristic parameters robust. The stickiness and universality are not high, which in turn affects the detection accuracy of material fatigue damage. Therefore, the magnetic multi-parameter fusion and fatigue state assessment are of great significance to ensure the safety and reliable operation of major mechanical equipment. The report introduced the observation uncertainty caused by the randomness of the MBN signal, and the influence of the model uncertainty caused by the model type, input, model parameter setting and training on the material fatigue evaluation, forming a magnetic field based on the observation uncertainty analysis. The multi-parameter fusion method analyzes the influence law of model uncertainty, and summarizes the method to improve the accuracy of the fatigue state assessment model.

Li Xiang, PhD, associate professor. Working in University of Electronic Science and Technology of China, School of Aeronautics and Astronautics. Engaged in the research of non-destructive testing methods, the main research directions include structural fatigue state testing and evaluation, stress testing, defect testing, etc., supervisory inspection technology based on digital twins, and corresponding data processing methods.

Undertake 5 various national scientific research projects, including the National Natural Science Foundation of China, a major project of a certain ministry, etc., and participate in key research and development projects as a backbone member. Won the third prize of Sichuan Science and Technology Progress Award, the first-class undergraduate course in Sichuan Province, published more than 20 academic papers, and participated in the formulation of 2 standards.