Open AccessA Bayesian estimation-based uncertainty quantification of flaws in steel welds detected by ultrasound phased array
Author(s) -
Jingran He,
Zhijian Chen,
Xiangyu Luo
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1592/1/012011
Subject(s) - phased array , welding , bayesian probability , nondestructive testing , uncertainty quantification , ultrasonic testing , scale (ratio) , ultrasonic sensor , acoustics , computer science , engineering , artificial intelligence , mechanical engineering , machine learning , physics , antenna (radio) , telecommunications , quantum mechanics
Phased array ultrasonic scanning is currently the most advanced non-destructive test for detecting flaws in a steel welding. However, large uncertainty exists in the detection results. Technically, this uncertainty is induced by the sparsity of detection resolution compared to the scale of the flaws. Consequently, it is essential to enrich the resolution of detection results and quantify the corresponding statistical uncertainty before evaluating the number and equivalent of the flaws in the steel welding. In this paper an uncertainty quantification framework is built with Bayesian compressive sensing method from which the equivalent of the flaws in the steel welding with certain degree of confidence can finally be given.