Kangwei Wang

Dr. Kangwei Wang

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 June 2021 in Structural Health Monitoring
Verified
Journal Article
Closed access
PDF file
Kangwei WangKangwei Wang
Kangwei Wang
JZJie Zhang
Jie Zhang
YSYi Shen
Yi Shen
BKBenjamin Karkera
Benjamin Karkera
ACAnthony J Croxford
Anthony J Croxford
PWPaul D Wilcox
Paul D Wilcox
To perform long-term structural health monitoring, a method based on a nonlinear autoregressive exogenous network is used to learn the features present in signals acquired from a pristine structure. When a subsequent measured signal is input to the trained nonlinear autoregressive exogenous network, the output is a prediction of the equivalent signal from a pristine structure. The residual when the pristine predicted signal is subtracted from the measured signal is used for defect detection and localization. A methodology of how to train, test and assess a nonlinear autoregressive exogenous network for guided wave signals is introduced and applied to experimental data obtained over a period of 8 years from a sparse array of guided wave sensors deployed on a steel storage tank. A separate nonlinear autoregressive exogenous model is trained for each sensor pair in the array using data captured in 2012. The method is first tested using data from a single pair of sensors. Defect signals are synthesized by superposing simulated responses from defects onto later experimental signals obtained from the real structure. The test results for the nonlinear autoregressive exogenous method show better detection performance than those from the optimal baseline selection method, in terms of receiver operating characteristic curves. The detection performance of the nonlinear autoregressive exogenous method is further assessed on signals from the whole sensor array, again with simulated defect responses superposed. It is shown that good detection and localization performance can be achieved by combining the nonlinear autoregressive exogenous residual signals from different sensor pairs. The nonlinear autoregressive exogenous method is tested on experimental data acquired at intervals over the following 7 years as the condition of the tank naturally degrades. Indications from localized corrosion are observed. Finally, an artificial localized anomaly is added to the tank and is visible at the correct location in the image formed using the nonlinear autoregressive exogenous method.

9 October 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Kangwei WangKangwei Wang
Kangwei Wang
XZXin Zhang
Xin Zhang
FWFengshuo Wan
Fengshuo Wan
RCRong Chen
Rong Chen
JZJun Zhang
Jun Zhang
Yong YangYong Yang
Yong Yang
Wheel defects, such as flats, can have severe impacts on vehicles, seriously hindering railway running stability and safety. Thus, it is crucial to continuously monitor the condition of wheelsets using techniques such as acoustic emission (AE). In this article, we propose an attentive feature selection sequential (AFSS) module, inspired by attention-based mechanisms, which can be integrated into deep sequential neural networks, such as bidirectional gated recurrent unit. The feature sets were extracted from the AE signals acquired with various wheel defects using Mel-frequency cepstral coefficients and Gammatone cepstral coefficients. Next, irrelevant and redundant features were removed and the optimal feature subset was obtained with the simultaneous training of the AFSS modules and trunk classifier network. The proposed method was compared with several state-of-the-art feature selection algorithms. The evaluation was conducted using experimental wheel defect datasets. Considering the imbalance in the datasets, the methods were further enhanced using a ROC curve-based threshold rescaling strategy. The experimental results demonstrate that AFSS outperforms other algorithms in terms of accuracy and generalization without an excessive increase in computational costs. Moreover, AFSS exhibits superior stability and noise resistance, as confirmed by false-alarm rate tests in practical defect-free validations. Overall, the proposed method combines the benefits of the learning capabilities of sequential neural networks and embedding of feature selection in classifiers. Thus, this study provides theoretical guidance for employing feature selection techniques in sequential network modelling and wheel defect detection tasks.

12 November 2021 in Applied Artificial Intelligence
Verified
Journal Article
Closed access
Kangwei WangKangwei Wang
Kangwei Wang
XZXin Zhang
Xin Zhang
SSShuzhi Song
Shuzhi Song
YWYan Wang
Yan Wang
YSYi Shen
Yi Shen
PWPaul D. Wilcox
Paul D. Wilcox
In order to improve the effectiveness of the acoustic emission (AE) technique in rail health monitoring, a novel genetic clustering technique is proposed to categorize data automatically, integrating density-based clustering and t-distributed stochastic neighbor embedding. A primary problem in optimizing density-based clustering is to accommodate noise, for it explicitly computes the noise subset. Thus, the generalized silhouette index is proposed as a profitable objective to properly tackle noise and arbitrary shapes. The proposed method is initially testified in ten benchmark datasets, which manifests a superiority in handling irregular shape datasets and noise interference. Furthermore, the proposed method is applied in real-world AE signals acquired from tensile tests. The clustering results elucidated that it outperforms the comparative methods in categorizing the fused AE features and remains robust with increasing railway noise interference. In conclusion, the proposed method is validated to discover intrinsic groups of AE data and analyze potential rail health stages.

17 November 2021 in Measurement Science and Technology
Verified
Journal Article
Closed access
Kangwei WangKangwei Wang
Kangwei Wang
XZXin Zhang
Xin Zhang
SSShuzhi Song
Shuzhi Song
YWYan Wang
Yan Wang
YSYi Shen
Yi Shen
: In an attempt to analyze the feasibility of a rail monitoring scheme with a wheel-mounted acoustic emission measurement method, a two-dimensional analytical spring model was proposed to interpret the interactions between acoustic waves and stationary wheel–rail contact interfaces. The spring model represents the coupling strength of the interfaces with their stiffness. It accommodates the environmental information of contact stiffness and the acoustic source information, such as the feature frequency and the incident angle in the analysis of interface transmissibility. The fractal dimension of the interface is introduced into the spring model to interpret the effect of the axle load on the interface stiffness and the variation in transmissibility. Discrete Rayleigh integration is further combined to obtain the directivity of the interface and estimate the acoustic field intensity in the entire wheel. A wheel–rail contact rig was designed to simulate the actual contact conditions in a railway. Experimental data acquired from this test rig were utilized to validate the model, with regard to the interfacial stiffness in normal incidence and amplitudes of the transmitted waves in oblique incidence. After verifying the reliability of the model, the acoustic field intensity in the wheel was visualized under the assumed environmental conditions. Finally, a discussion is presented to determine a proper angular separation for the wheel-mounted sensor system, and substantiate the feasibility of the system.

15 November 2024 in Applied Acoustics
Verified
Journal Article
Open access
Kangwei WangKangwei Wang
Kangwei Wang
YQYang Qian
Yang Qian
JXJie Xie
Jie Xie
JWJun Wang
Jun Wang
WHWeiguo Huang
Weiguo Huang

7 March 2024 in Structural Health Monitoring
Verified
Journal Article
Closed access
Kangwei WangKangwei Wang
Kangwei Wang
JZJie Zhang
Jie Zhang
YXYang XIAO
Yang XIAO
ACAnthony J. Croxford
Anthony J. Croxford
YYYong Yang
Yong Yang
Guided wave structural health monitoring (GWSHM) systems, using the delay-and-sum imaging algorithm, are an efficient solution to detect and localise defects in industrial structures. However, the image artifacts caused by either imperfect detection or sensor lay-out limitations make it difficult to identify and locate defects accurately. In order to enhance the performance of defect detection and localisation in GWSHM systems, this paper proposes a three-step procedure for post-processing guided wave signals prior to image formation. In the first step, the signals are processed using the nonlinear autoregressive exogenous model to suppress noise from benign features. The second step calculates the probability of defect presence based on the rescaled Gamma cumulative distribution function. This probabilistic threshold is then determined from the quantile mapping. Finally, a guide wave image is formed using the delay-and-sum imaging algorithm. The experimental validation was performed to inspect a 6 mm-diameter through-thickness circular hole on an aluminium plate and the defects were further scaled as simulated datasets to test its detectability under various amplitudes. In the second procedure step, the detection and localisation performance of the proposed procedure was compared with that of using the signal difference coefficient and the Rayleigh maximum likelihood estimator. It is shown that the proposed procedure can enhance the contrast between damaged and undamaged regions, providing more reliable and accurate guided wave images.