JOINT DAMAGE IDENTIFICATION FOR A STEEL FRAME USING COMPRESSION SENSING-BASED EMI SIGNATURES
An experimental study of damage identification for a three-layered steel frame is implemented.Damages are introduced by completely loosening bolts over several conditions on the structure and then the piezoelectric admittance signatures are obtained under various damage locations.EMI signatures are pre-processed using the theory of compression sensing and the compressed signatures still keep the original damage information invariable.Finally, the principal components of the compressed EMI signatures are obtained as the input parameters of BP neural network for quantitative damage identification.Research shows that the transmission bandwidth and storage space of the processed data are reduced greatly using the compression perception technique, and the present hybrid method can detect the damage locations quantitatively.
Damage identification Three-layered frame Piezoelectric admittance Compression sensing Principal component analysis (PCA) BP neural network
Tian-bi LV Wei YAN Bi-quan PENG
Faculty of Architectural, Civil Engineering and Environment, Ningbo University, Ningbo, P.R.China, 315211
国际会议
The 2017 Symposium on Piezoelectricity,Acoustic Waves and Device Applications(2017全国压电和声波理论及器件技术研讨会)
成都
英文
226-229
2017-10-27（万方平台首次上网日期，不代表论文的发表时间）