A multilevel Bayesian method for ultrasound-based damage identification in composite laminates

详细信息    查看全文

• 作者：Juan Chiach&iacute ; o ; ^{jchiachio@ugr.es} ; Nicolas Bochud ; Manuel Chiach&iacute ; o ; Sergio Cantero ; Guillermo Rus
• 关键词：Bayesian inverse problem ; Ultrasound ; Composite laminates ; Diagnostics
• 刊名：Mechanical Systems and Signal Processing
• 出版年：2017
• 出版时间：1 May 2017
• 年：2017
• 卷：88
• 期：Complete
• 页码：462-477
• 全文大小：1534 K
• 卷排序：88

文摘

A multilevel Bayesian approach is proposed for damage assessment in layered materials. The method does not require any predefined hypothesis about damage distribution. Several damage hypotheses are tested and ranked through relative probabilities. A case study is presented using through-transmission ultrasonic data in a CFRP laminate. The method enforces a quantitative Ockham's razor for damage hypothesis selection.