Damage detection using vibration data and dynamic transmissibility ensemble with auto-associative neural network

Output-Based Structural Damage Detection by Using Correlation Analysis Together with Transmissibility

“…The proposed methodology for inspecting structural damage is calculated for the cantilever beam, and the results are drawn hereafter. In this section, the transmissibility is derived; unlike the previous investigations [ 10 , 11 , 12 , 13 , 14 , 15 ], transmissibility is selected upon a specified reference. This study uses ab average transmissibility with a changing reference for all available cases; this is the same in the further experiment investigation.…”

“…In order to solve this problem, output-based techniques are thus developed; so-called operational modal analysis (OMA). Among all vibration-based techniques, transmissibility occupies a significant role as it only relies on the structural responses, without any requirement on the excitation measurement [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ]. Transmissibility has been shown to perform well in damage detection, and is applied to damage detection [ 8 , 9 , 12 , 13 , 14 ], localization [ 10 ], and quantification [ 10 , 11 , 15 ].…”

“…[ 8 ], principal component analysis (PCA) is applied to reduce the dimension of transmissibility, then the distance measure is applied to enhance damage detection performance; in ref. [ 9 ], transmissibility works together with Mahalanobis distance and Hotelling T square for damage detection; in [ 11 ], a new concept—transmissibility coherence—is raised and applied to damage detection and quantification; transmissibility can also work with distance measuring [ 12 ], clustering analysis [ 13 ], neural networks [ 14 ], and similarity measuring [ 15 ] in identifying the structural damages.…”

“…Certainly, it is also possible to consider factors such as corrosion, fatigue loading [ 21 ], and so on as damage. It is also feasible to consider the consequences of the damage, such as stiffness change [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ] or damping change [ 22 ]. Accordingly, SHM methodologies can be chosen from different perspectives.…”

Structural Damage Identification Based on the Transmissibility Function and Support Vector Machine

“…Chen et al [11] tried to predict damage with neural networks (NNs) trained by TF; the test results showed that the NN classifiers clearly delivered the diagnostic indications of the faults introduced into the structural systems. In [12], a TF based damage detection method using artificial intelligence was proposed for detecting the structural damage and predicting its severity.…”