Mono-modal feature extraction for bonding quality detection of explosive clad structure with optimized dual-tree complex wavelet transform

A novel method for bonding state detection of explosive clad structure based on the Hilbert envelop energy ratio of mono-model signals

“…It can be known from equation (24) that the Hilbert envelop energy ratio Y of arbitrary two order model signals is only related with intrinsic property parameters damping ratio ζ, undamped natural frequency ω and principal mode f of the system, and unrelated with impulse force and other external conditions. For explosive clad structure, through a great deal of experiments and theory studies in the authors' previous research [25][26][27], it has been concluded that damping ratio is sensitive to the change of the bonding state. With the increasing of the bonding strength, the high order modal damping ratio is reduced and the low order modal damping ratio has an increased trend.…”

“…In order to validate the proposed method, it is applied to bonding state detection of explosive clad pipe, and compared with the detection method based on ODTCWT and time entropy index which is proposed by the authors in previous research [27,28]. The experimental results show the robustness and effectiveness of the proposed method.…”

“…Our lab has carried out the research about the bonding state detection of explosive clad structure based on this idea. The authors have used ensemble empirical mode decomposition [25], dual-tree complex wavelet transform [26] and optimized dual-tree complex wavelet transform (ODTCWT) [27] to extract the feature information from vibration response signal, and then constructed the time entropy as detection index to identify the bonding state of such structure. However, the time entropy index is not sensitive enough to the small change of the bonding state.…”

“…Generally, the Hilbert envelope energy ratio of arbitrary two order model signals from impulse vibration response is only related with the modal parameters such as damping ratios, natural frequencies and principal modes of MDOF system, and it can reflect the inherent property of MDOF system. For explosive clad structure, through a great deal of experiments and theory studies in the authors' previous research [25][26][27], it has been concluded that natural frequencies are not sensitive to the change of the bonding state, but damping ratio is sensitive to the change of the bonding state. With the increasing of the bonding strength, the high order modal damping ratio was reduced and the low order modal damping ratio had a increased trend.…”

Looseness condition feature extraction of viscoelastic sandwich structure using dual-tree complex wavelet packet-based deep autoencoder network

“…Dual-tree complex wavelet transform (DTCWT), with the properties of reduced spectral aliasing, nearly shift invariance, and limited data redundancy, has been widely used in fault feature extraction of mechanical equipment. 10,11 DTCWT only focuses on the low-frequency band, which may leave out useful features in the high-frequency band. In order to extract effective component sufficiently in the whole band, dual-tree complex wavelet packet transform (DTCWPT) is used to decompose the vibration response signal of VSS for looseness condition feature components extraction in this article.…”