Ultrasonic Guided Wave Damage Localization in Hole-structural Bearing Crossbeam Based on Improved RAPID Algorithm

• Published in: IEEE transactions on instrumentation and measurement 2022, Vol.71, p.1-1
• Main Authors: Teng, Feiyu, Wei, Juntao, Lv, Shanshan, Peng, Chang, Zhang, Lei, Ju, Zengye, Jia, Lei, Jiang, Mingshun
• Format: Article
• Language: English
• Subjects: Algorithms; Computer simulation; Correlation analysis; Crossbeam; Damage localization; Distribution functions; Fatigue cracks; High speed rail; Image reconstruction; Inspection; Load modeling; Localization; Location awareness; Mathematical analysis; Monitoring; Numerical models; Performance degradation; Probability distribution; probability distribution function; Probability distribution functions; Shape; Shape factor; Statistical analysis; Stress; Stress distribution; Structural damage; Structural health monitoring; time of flight; ultrasonic guided wave
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2022.3207808

The hole-structural bearing crossbeam plays an important role in high-speed trains. However, in the service process, the long-term fatigue load may lead to structural damage such as cracks, resulting in performance degradation and failure. Ultrasonic guided wave technology is one of the most effective damage localization methods in structural health monitoring (SHM) with a high damage sensitivity and wide monitoring range. To address the damage localization in bearing crossbeams, a modified reconstruction algorithm for probabilistic inspection of damage (RAPID) based on corrected probability distribution function is proposed. Firstly, the valid sensor paths affected by damage are obtained using damage index (DI) based on correlation analysis. Then, the positional relationships between valid paths and damage are classified based on the time of flight (TOF). Finally, the damage diagnostic image and localization are obtained by fusion imaging using the corresponding probability distribution functions and shape factors, depending on the different types of the path. The effectiveness was verified by numerical simulation and experiment. Taking the crossbeam of high-speed train as the research object, through the static simulation of the crossbeam stress distribution under load, the damage hot-spots area is obtained, and the sensor network is designed. And then, the SHM experimental system is constructed to perform damage localization experiments. The localization absolute error was less than 8 mm. Experimental results show that the proposed method can effectively locate the damage position in the crossbeam, and has better accuracy and reliability than the traditional RAPID algorithm.