文档介绍：第28卷第3期 计算机集成制造系统 ters of linear time-invariant three- dimensional structures only
fromthestationaryvibrationresponsesignals,
responsedatawasregardedasthedatasetinthehigh-dimensionalspace, andthelow-dimensionalembeddeddata
, thenaturalfrequencyofthe modalwasidentifiedfromthe modal
responsematrixcorrespondingtothelowdimensionalembedded matrixbyusingthesingledegreeoffreedomrecog-
modal shapes could be calculated from the Moore-Penrose matrixinverseofthelow-dimen-
-dimensionalcylindricalshelshowedthatcomparedwithe-
quidistantmapping, Laplacian Eigenmaps wasefectivelyidentifythe modalshapeand naturalfrequency ofthe
-
sis, LaplacianEigenmapshadhigherrecognitionaccuracy.
收稿日期：2020-04-14；修订日期：20201018。Received 14 Apr. 2020； accepted 18 Oct. 2020.
基金项目：国家自然科学基金资助项目(51305142,51505145)福建省科技计划引导性资助项目(2017H01010065 #021H0019)；中国博士后科
学基金第55批面上资助项目(014M552429)泉州市科技计划资助项目(018C110R,2018C114R