Fault tracing in multistage gearbox systems based on an improved transfer path analysis method
Penghong Lu, Cai Li, Gang Chen, Gang Chen*
Measurement,242,2025
PDF   Abstract   BibTeX  

This study, has developed a novel structural dynamics decoupling method based on TPA. This approach effectively facilitates fault diagnosis and tracing in multistage gearbox systems. Initially, the research explores the relationship between the transfer function of systems with rigid link decoupling and the coupled system’s frequency response function. Compared with Huangfu et al. (2023) , the signal measurement points are reduced. The MWI (Yu et al., 2023) is employed to solve the inverse problem of bearing force identification. Subsequently, the decoupled frequency response and bearing forces are multiplied to calculate the path contribution, determining the dominant fault transfer path and thereby facilitating fault tracing in gearboxes. The effectiveness of this method is validated through numerical simulations and experimental studies, with fault characteristic enhancement exceeding 200%. To evaluate the …

              @article{chen2024enhancing,
                title={Enhancing Reliability through Interpretability: A Comprehensive Survey of Interpretable Intelligent Fault Diagnosis in Rotating Machinery},
                author={Chen, Gang and Yuan, Junlin and Zhang, Yiyue and Zhu, Hanyue and Huang, Ruyi and Wang, Fengtao and Li, Weihua},
                journal={IEEE Access},
                year={2024},
                publisher={IEEE}
              }
            

VNCCD: A gearbox fault diagnosis technique under nonstationary conditions via virtual decoupled transfer path
Cai Li, Penghong Lu, Gang Chen*
Mechanical Systems and Signal Processing,221,2024
PDF   Abstract   BibTeX  

The interconnection between different components among a gearbox poses a challenge in decoupling vibration signals for fault diagnosis. This study presents a method called Virtual Non-Linear Chirp Component Decoupling (VNCCD) to address gearbox fault diagnosis. The method starts by identifying the decoupled frequency response function (DFRF) using a virtual decoupling method. The bearing force signals are then computed based on the DFRF, and the intrinsic signal components are extracted using a nonlinear chirp mode decomposition technique. The proposed approach enhances the quality of vibration signals in gear systems by eliminating the intercoupling effect of structural transfer paths. By employing the VNCCD method, the amplification of fault conditions can be effectively demonstrated compared to the nonlinear chirp component obtained from the original signal. Simulation and experimental …

              @article{chen2024enhancing,
                title={Enhancing Reliability through Interpretability: A Comprehensive Survey of Interpretable Intelligent Fault Diagnosis in Rotating Machinery},
                author={Chen, Gang and Yuan, Junlin and Zhang, Yiyue and Zhu, Hanyue and Huang, Ruyi and Wang, Fengtao and Li, Weihua},
                journal={IEEE Access},
                year={2024},
                publisher={IEEE}
              }
            

Bearing Fault Diagnosis via Robust PCA with Nonconvex Rank Approximation
Cai Li; Penghong Lu; Guangming Dong, and Gang Chen*
IEEE Sensors Journal,2024
PDF   Abstract   BibTeX  

Feature extraction is an essential part of bearing fault diagnosis. Robust principal component analysis (RPCA) provides a general technique for extracting fault features. However, Although the convex relaxed RPCA is convex and easy to optimize, the global optimal solution obtained may deviate significantly from reality. To address this issue, we propose a nonconvex RPCA method for bearing fault diagnosis, utilizing the γ norm penalty. Leveraging the unitary invariance property of the γ norm and the Moreau-Yosida operator, we transform the original equation into a singular value optimization problem using a difference of convex programming to obtain a feasible solution. The most important advantage of Nonconvex RPCA over conventional filtering methods is that it can improve fault feature extraction while decreasing noise interference, which results in substantially enhanced estimation accuracy of the bearing fault signal. Simulation analysis and experimental results confirm the effectiveness of the developed approach. Comparison experiments demonstrate that nonconvex RPCA provides more accurate extraction results compared to L1-norm regularization, Variational Mode Decomposition, and Feature Mode Decomposition.

              @article{li2024bearing,
                title={Bearing Fault Diagnosis via Robust PCA with Nonconvex Rank Approximation},
                author={Li, Cai and Lu, Penghong and Dong, Guangming and Chen, Gang},
                journal={IEEE Sensors Journal},
                year={2024},
                publisher={IEEE}
              }
            

Chirplet Wigner–Ville distribution for time–frequency representation and its application
Gang Chen, Jin Chen, Guangming Dong
Mechanical Systems and Signal Processing,2013
PDF   Abstract   BibTeX    

This paper presents a Chirplet Wigner–Ville Distribution (CWVD) that is free for cross-term that usually occurs in Wigner–Ville distribution (WVD). By transforming the signal with frequency rotating operators, several mono-frequency signals without intermittent are obtained, WVD is applied to the rotated signals that is cross-term free, then some frequency shift operators corresponding to the rotating operator are utilized to relocate the signal′s instantaneous frequencies (IFs). The operators′ parameters come from the estimation of the IFs which are approached with a polynomial functions or spline functions. What is more, by analysis of error, the main factors for the performance of the novel method have been discovered and an effective signal extending method based on the IFs estimation has been developed to improve the energy concentration of WVD. The excellent performance of the novel method was …

                @article{chen2013chirplet,
                  title={Chirplet Wigner--Ville distribution for time--frequency representation and its application},
                  author={Chen, G and Chen, J and Dong, GM},
                  journal={Mechanical Systems and Signal Processing},
                  volume={41},
                  number={1-2},
                  pages={1--13},
                  year={2013},
                  publisher={Elsevier}
                }