Sinian Wang

Sinian Wang

Auto-generated profile, based on publicly available information

Sinian Wang

College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
13 September 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
LFLei Fu
Lei Fu
SWSinian Wang
Sinian Wang
RCRuien Chen
Ruien Chen
ZMzepeng ma
zepeng ma
JMJianhui Ma
Jianhui Ma
FXFang XU
Fang XU
Rolling bearing failures can lead to machinery breakdowns and loss of control, posing significant operational and safety risks. Traditional algorithms based on the empirical wavelet transform (EWT) face challenges in accurately extracting valid components under variable conditions. Additionally, conventional maxpooling methods often overlook critical features, while fully connected layers tend to overfitting due to excessive parameterization. To address these challenges, an enhanced algorithm is proposed, defined as the improved EWT-improved ResNet (IEWT-IResNet) method. Firstly, the vibration signals are preprocessed to enhance resilience under variable conditions. Secondly, an improved EWT (IEWT) is proposed with an optimal ridge-curve extraction strategy. And it incorporates an efficient weighted energy entropy (EWEE) as a sensitive indicator for signal reconstruction. Additionally, the 1D angular-serial signals are transformed into 2D image matrices using the generalized S-transform. Finally, an optimized residual network model is presented, incorporating singular value decomposition (SVD) pooling and global SVD pooling (GSP) strategies in the pooling and full connection layers. These enhancements benefits the optimal feature extraction while mitigating overfitting. The experimental results demonstrate the superior feature extraction capabilities of the proposed method when processing noisy and nonstationary signals. Furthermore, the effectiveness of the approach is investigated and validated using real-world data from rolling bearings.

16 November 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
LFLei Fu
Lei Fu
PZPengshuai Zhang
Pengshuai Zhang
MDMeiya Ding
Meiya Ding
SWSinian Wang
Sinian Wang
FXFang XU
Fang XU
ZMzepeng ma
zepeng ma
Traditional deconvolution methods struggle to separate fault information under interference from contaminative signals. Although some improved efforts tried, their application is strictly limited to parameter requirements. To overcome this, a composite envelope negentropy deconvolution and reconstruction (CENDR) method is proposed. First, considering the inherent sensitivity to repetitious transient features in the energy domain, composite envelope negentropy (CEN) is utilized as the objective function for deconvolution to enhance the fault features. Then, an adaptive reconstruction method based on CEN and singular value decomposition (SVD) is proposed to reconstruct the filtered signal. The superiority of CENDR lies in its ability to simultaneously consider the periodicity and impulsiveness of fault information. It achieves fault feature extraction under strong interference without relying on strict parameters. Specifically, by utilizing fault characteristic ratio (FCR), the quantitative evaluation reveals that CENDR performs better than the existing optimal deconvolution algorithm, ranging from 28% to 404% under different scenarios.