Non-contact modal parameters identification using a K-cluster algorithm

摘要

Non-contact structural health monitoring is a promising field for assessing civil structures, such as bridges. Not having toacces s th e structur e avoid s differen t issues : th e closur e o f th e structure , th e us e o f specia l equipmen t t o acces s i t, andothers . This study uses digital image processing , machine learning , and paralle l computing to detec t th e vibration o f aflexible structure. If a structure is too stiff, a reinforced concrete short-span bridge or a multi-story building, it is hard toidentify its natural frequencies without some sort of target panel or target feature. Instead, if the structure is flexible, it ispossible to identify its displacement and its natural frequencies, but it is a challenge with high computational cost. Thisstu dy prese nts an unsupervis ed machine-learni ng algorith m to ident ify a structu re, its displaceme nt, a nd its naturalfrequencie s. Th e algorithm wa s deploye d o n a simpl e supporte d beam usin g a commerciall y availabl e camer a an d aninexpensive GPU.