Jian Cheng

Jian Cheng

Auto-generated profile, based on publicly available information

Jian Cheng

School of Mechanical Engineering, Anhui University of Technology, Ma’anshan, China

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
8 November 2021 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
Xin LiXin Li
Xin Li
JCJian Cheng
Jian Cheng
Haidong ShaoHaidong Shao
Haidong Shao
Kan LiuKan Liu
Kan Liu
Baoping CaiBaoping Cai
Baoping Cai
Vibration signals and infrared images have different advantages and characteristics. Although a few recent researches have explored their information fusion in rotating machinery fault diagnosis, they show limited performance when facing strong interference and imbalanced cases. Therefore, a fusion framework based on confidence weight support matrix machine (CWSMM) is proposed. In this framework, CWSMM can not only fully leverage the structure information of infrared thermography images and vibration time–frequency images, but also has the following novelties. First, CWSMM uses dynamic penalty factors for different class samples to address the class imbalance problem. Second, by using the prior knowledge of matrix samples, a confidence weight assignment strategy is designed for CWSMM to improve the robustness. Last, the Dempster–Shafer (D-S) evidence theory is applied to fuse the posterior probability outputs of CWSMMs using different measurements. Experiment results demonstrate that the proposed method has promising fault diagnosis performance, specifically under strong interference and imbalanced datasets.

3 December 2021 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
JCJian Cheng
Jian Cheng
YYYu Yang
Yu Yang
ZWZhantao Wu
Zhantao Wu
Haidong ShaoHaidong Shao
Haidong Shao
HPHaiyang Pan
Haiyang Pan
JCJunsheng Cheng
Junsheng Cheng
As an important part of rotating machinery, gear is easy to appear some unexpected fault states, and its fault diagnosis is very important. Fourier decomposition method (FDM) is a common method for gear fault diagnosis, but the noise robustness, period recognition, and extraction capabilities of FDM are unsatisfactory. Based on this, in this article, Ramanujan Fourier mode decomposition (RFMD) method is proposed. The RFMD not only has a complete mathematical theory foundation but also has an excellent ability to identify and extract periodic components. Emulational and experimental results of planetary gearbox show that the RFMD method has good noise robustness and can accurately extract gear fault characteristic information. Thus, it is an effective gear fault diagnosis method.

24 June 2024 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
JCJian Cheng
Jian Cheng
Haiyang PanHaiyang Pan
Haiyang Pan
Jinde ZhengJinde Zheng
Jinde Zheng
In this article, a new deconvolution method, named maximum Ramanujan spectrum signal-to-noise ratio deconvolution (MRSD) method is proposed. MRSD updates the filter by maximizing the index of v -Ramanujan spectrum signal-to-noise ratio ( v -RSSNR) to improve the noise reduction effect and the performance of feature enhancement. On the one hand, the concept of generalized envelope is introduced into the MRSD method, and flexible envelopes are used to enhance the weak state features, and the v -Ramanujan spectrum of the signal is analyzed by using the mixed Ramanujan Fourier transform, so as to provide an optimal plane for the evaluation of weak state features. On the other hand, the MRSD method designs the filter by maximizing the v -RSSNR index, and optimizes the objective function by gradient descent. The simulation and experimental analysis results show that MRSD method is an effective noise reduction method and can accurately extract weak state features.

18 March 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
JCJian Cheng
Jian Cheng
Haiyang PanHaiyang Pan
Haiyang Pan
JTJinyu Tong
Jinyu Tong
Jinde ZhengJinde Zheng
Jinde Zheng
Although traditional signal processing methods have good decomposition performance in multimodal signals, they lack theoretical research on periodic pulse signals, resulting in insufficient decomposition. Based on this, a reduced mode decomposition (RMD) method is proposed in this paper, which can decompose reduced components (RCs) iteratively through the designed finite impulse response (FIR) filter bank. On the one hand, an adaptive index called reweighted kurtosis (RK) is defined as the objective function of filter banks, so as to fully consider the impulsivity and periodicity of signals and make the filter components contain rich state information. On the other hand, FIR is used to constrain various modal signals, so that RMD is not limited by filter bandwidth and center frequency, improves the decomposition ability and ensures the robustness of noise. The verification results of two types of roller bearing fault signals indicate that RMD is a novel multi-mode signal analysis method.

10 October 2023 in Structural Health Monitoring
Verified
Journal Article
Closed access
Haiyang PanHaiyang Pan
Haiyang Pan
HFHong Feng
Hong Feng
JCJian Cheng
Jian Cheng
JZJinde Zheng
Jinde Zheng
Under the influence of strong noise, period fault features of rolling bearing are not obvious, which increases the difficulty of accurately extracting period fault features. An autonomous weak period fault extraction method under strong noise named Ramanujan-gram is proposed in this paper. The greatest advantage of Ramanujan-gram is that it uses the Ramanujan feature extraction technique to reconstruct the components in each frequency band, which can overcome the weakness of the weak noise robustness of the filter methods used by the traditional kurtogram methods and improve the accuracy of period fault feature extraction. Meanwhile, the adaptive frequency band segmentation method based on the order statistical filter is used for adaptive frequency band segmentation, which overcomes the defect that the binary tree structure of fixed frequency band segmentation may destroy the optimal demodulated frequency band. Considering that kurtosis index is difficult to accurately evaluate period fault information in components, Ramanujan-gram adopts adaptive square envelope spectrum weighted kurtosis index to improve the evaluation accuracy of period fault information. The test signals of rolling bearing verify that Ramanujan-gram has strong noise robustness and is an effective method for weak period fault extraction under strong noise.

27 April 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Haiyang PanHaiyang Pan
Haiyang Pan
YZYing Zhang
Ying Zhang
JCJian Cheng
Jian Cheng
Jinde ZhengJinde Zheng
Jinde Zheng
Symplectic geometry mode decomposition (SGMD) method takes the Hankel matrix as the trajectory matrix, and the eigenvalue of trajectory matrix can be obtained by symplectic geometry similarity transformation (SGST). However, with the increase of noise intensity, SGMD method based on Hankel matrix cannot distinguish the fault signal excited by defects and background noise at the same order of magnitude. Based on this, a symplectic geometry transformation based periodic segment (SGT-PS) method is proposed. In SGT-PS, a neighboring peak method is designed to estimate the signal period and determine the variable parameters, which overcomes the defect that SGMD is difficult to extract the periodic pulse components. Meanwhile, optimized periodic segment matrix (OPSM) is defined to segment periodic pulse information, reduce the accumulated error and improve the accuracy of periodic pulse extraction. The analysis results of roller bearing fault signals show that SGT-PS is an effective signal decomposition method, which can accurately extract the periodic pulse.