Article
Acoustics
Ruiqi Zhang, Liang Guo, Zhuyuxiu Zong, Hongli Gao, Mengui Qian, Zaigang Chen
Summary: Accurately establishing a dynamic model of rolling bearings is essential for understanding the fault mechanism and analyzing the motion characteristics. A four-degree-of-freedom bearing dynamics model based on Hertz basis theory is proposed, considering the time-varying displacement excitation function and validating its accuracy through comparison with experimental signals. The influence of rotational speeds and defect sizes on the vibration characteristics of the bearing rolling element fault is analyzed and summarized based on the mechanical model and experimental data.
JOURNAL OF SOUND AND VIBRATION
(2023)
Article
Acoustics
T. Haj Mohamad, C. Nataraj
Summary: This article presents a new method for diagnosing rolling element bearings in rotating machines using phase space topology and time-domain statistical features. The results show high accuracy, recall, and precision of this method, with superior performance compared to traditional envelope analysis methods.
JOURNAL OF VIBRATION AND CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Dawei Gao, Yongsheng Zhu, Zhijun Ren, Ke Yan, Wei Kang
Summary: Traditional diagnostic methods are less accurate for bearings under strong noise conditions, making the extraction of weak fault features a research focus. This paper proposes a novel method consisting of multi-channel continuous wavelet transform and convolution-feature-based recurrent neural network, which effectively improves the diagnostic efficiency for bearings.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Engineering, Multidisciplinary
Jiahui Tang, Jimei Wu, Bingbing Hu, Jie Liu
Summary: The proposed method for bearing fault diagnosis based on fault feature region detection utilizes a deep belief network to train the proposed regions with fault features, achieving a prediction accuracy of over 80% for compound faults.
Article
Engineering, Multidisciplinary
Ganesh L. Suryawanshi, Sachin K. Patil, Ramchandra G. Desavale
Summary: The study shows that increasing the inclination angle of surface faults significantly decreases the relative vibration amplitude at various rotor speeds but increases with fault depth. This empirically obtained variation in vibration amplitude can help predict the inclination of bearing surface faults.
Article
Engineering, Mechanical
Xin Chen, Yu Guo, Jing Na
Summary: The improved envelope spectrum via feature optimization-gram (IESFOgram) is an effective tool for improving the performance of the cyclic spectral coherence algorithm in revealing fault features of rolling element bearings (REB). In this study, an improved diagnostic feature (IDF) based on frequency tolerance and soft threshold denoising is proposed to address the issue of invalid demodulation bands for revealing fault characteristics. Simulation and experimental results verify the effectiveness of the proposed scheme.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Chemistry, Analytical
Zhengkun Xue, Yukun Huang, Wanyang Zhang, Jinchuan Shi, Huageng Luo
Summary: This paper proposes a novel fault feature extraction method that utilizes bidirectional composite coarse-graining process and fuzzy dispersion entropy, combined with random forest and maximum relevance minimum redundancy algorithm for feature selection. The effectiveness of the proposed method is evaluated through numerical simulation and experimental validation, demonstrating its capability of identifying fault categories and health states of rolling bearings simultaneously.
Article
Engineering, Electrical & Electronic
Mingliang Cui, Youqing Wang, Xinshuang Lin, Maiying Zhong
Summary: The study proposes a feature distance stack autoencoder (FD-SAE) for rolling bearing fault diagnosis, which has stronger feature extraction ability and faster network convergence speed compared to existing methods.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Yonghua Jiang, Zhuoqi Shi, Chao Tang, Jianfeng Sun, Linjie Zheng, Zengjie Qiu, Yian He, Guoqiang Li
Summary: This study proposes a new method called the dual domain adversarial network (DDAN) to address the issue of ineffective traditional diagnosis methods due to the feature distribution shift of rolling bearings under cross-conditions. The DDAN is integrated with a multichannel parallel feature extractor to extract features from both the frequency-domain and the time-frequency domain perspectives. The L(1,2 )norm based Wasserstein discrepancy ( L1,2 WD) is introduced to improve the stability and computation speed of the diagnosis model. A dual domain adversarial paradigm is constructed to improve the model's generalization by correcting overconfidence and expanding the confidence interval. The outcomes of verification on two bearing datasets demonstrate the excellence of DDAN in resolving cross-conditions rolling bearing fault diagnosis issues.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Review
Engineering, Chemical
Guangxi Yan, Jiang Chen, Yu Bai, Chengqing Yu, Chengming Yu
Summary: This paper reviews the current research status of rolling bearing fault diagnosis technology for railway vehicles, covering vibration, acoustic signal, and temperature prediction methods. The studies are compared and analyzed, and suggestions for improvement and possible research directions are proposed.
Article
Computer Science, Artificial Intelligence
Guanghua Fu, Bencheng Li, Yongsheng Yang, Chaofeng Li
Summary: This paper proposes a four-stage ensemble feature selection method called RTEFS, which can reduce data dimensionality and improve the accuracy and computational cost of machine learning models. Experimental results show that RTEFS outperforms the base counterparts in terms of accuracy and F-measure scores.
PATTERN RECOGNITION LETTERS
(2023)
Article
Acoustics
Huaitao Shi, Yangyang Li, Xiaotian Bai, Ke Zhang
Summary: This paper proposes a fault feature extraction method based on fusion signals by combining time-domain vibration and sound signals, effectively eliminating the decay of fault features in transmission paths and improving the performance of fault feature extraction.
Article
Engineering, Electrical & Electronic
Chaoang Xiao, Jianbo Yu
Summary: In this article, a novel adaptive swarm decomposition (ASWD) algorithm based on fine to coarse (FTC) segmentation is proposed for compound fault detection of rolling bearings. The algorithm automatically determines the number of oscillating components and employs the Teager energy kurtosis (TEK) as the indicator to evaluate the effectiveness of iterations. ASWD intelligently separates different oscillating components and suppresses redundant decomposition, effectively extracting compound fault impulses from multicomponent vibration signals. Experimental results show that ASWD outperforms other decomposition methods in terms of characteristic frequency intensity coefficient (CFIC).
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Hardware & Architecture
Gang Wang, Dongdong Liu, Lingli Cui
Summary: In this article, an auto-embedding transformer (AET) method is proposed to implement interpretable few-shot fault diagnosis of rolling bearings. The method improves the embedding quality of the signal and utilizes attention scoring to enhance the diagnostic accuracy and interpretability.
IEEE TRANSACTIONS ON RELIABILITY
(2023)
Article
Engineering, Electrical & Electronic
Wu Deng, Zhongxian Li, Xinyan Li, Huayue Chen, Huimin Zhao
Summary: This article proposes a novel compound fault diagnosis method MDSRCFD based on optimized MCKD and sparse representation, which separates and extracts the compound fault characteristics of rolling bearings through parameter optimization using intelligent optimization algorithms. The simulation and practical application results demonstrate that this method achieves accurate compound fault diagnosis.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)