A Robust Performance Degradation Modeling Approach Based on Student’s t-HMM and Nuisance Attribute Projection

Jiang, Huiming; Yuan, Jing; Zhao, Qian; Yan, Han; Wang, Sen; Shao, Yunfei

doi:10.1109/access.2020.2980019

Cited by 7 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So, HMMs are widely used in the field of mechanical fault diagnosis and PDA [ 31 , 32 ]. In this study, Student’s t-HMM, which has been proved to be highly tolerant to outliers in real-world applications [ 27 , 33 ], was introduced for bearing PDA based on the selected features. The graphical illustration of Student’s t-HMM is displayed in Figure 3 .…”

Section: Theory Backgroundmentioning

confidence: 99%

“…Yu et al [ 26 ] proposed an adaptive-learning-based method for machine faulty detection and health degradation monitoring, which provide a useful guide for developing a condition-based maintenance system. Jiang et al [ 27 ] combined Student’s t-HMM with nuisance attribute projection to construct a robust PDA model, which shows more tolerance to outliers than conventional HMMs. In this study, Student’s t-HMM was utilized to construct a health indicator based on the selected feature sets and to assess the degradation process.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bearing Performance Degradation Assessment Based on SC-RMI and Student’s t-HMM

et al. 2021

Self Cite

View full text Add to dashboard Cite

Bearing performance degradation assessment (PDA), as an important part of prognostics and health management (PHM), is significant to prevent major accidents and economic losses in industry. For the data-driven PDA, the extraction and selection of features is quite important. To better integrate the degradation information, the bearing performance degradation assessment based on SC-RMI and Student’s t-HMM is proposed in this article. Firstly, spectral clustering was used as a preprocessing step to cluster features with similar degradation curves. Then, rank mutual information, which is more suitable for trendability estimation of long time series, was utilized to select the optimal feature from each cluster. The feature selection method based on these two steps is called SC-RMI for short. With the selected features, Student’s t-HMM, which is more robust to outliers, was utilized for performance degradation modeling and assessment. The verifications based on an accelerated life test and the public XJTU-SY dataset showed the superiority of the proposed method.

show abstract

Section: Theory Backgroundmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Bearing Performance Degradation Assessment Based on SC-RMI and Student’s t-HMM

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Jiang et al [27][28][29] proposed a method by combining NAP and hidden Markov model (HMM) to evaluate the performance degradation of rolling bearings, and then they combined NAP with Student's t-hidden Markov model (Student's t-HMM) to obtain more accurate performance degradation assessments of rolling bearings. 30 Ma et al 31 imported NAP to the structure of convolutional neural network (CNN) to simplify the diagnosis process of rolling bearings under various speed conditions. Gandhi et al 32 introduced NAP to eliminate the interference of system dependent features and improve the robustness of fault diagnosis system of synchronous generators.…”

Section: Introductionmentioning

confidence: 99%

Robust fault diagnosis of rolling bearings via entropy-weighted nuisance attribute projection and neural network under various operating conditions

Yang

Yuan

et al. 2022

Structural Health Monitoring

View full text Add to dashboard Cite

Rolling bearings are crucial components in the fields of mechanical, civil, and aerospace engineering. They sometimes work under various operating conditions, which makes it harder to distinguish faults from normal signals. Nuisance attribute projection (NAP) is a technique that has been widely used in audio and image recognition to eliminate interference information in the extracted feature space. In constructing the weighted matrix of NAP, the setting of the weighted value represents the degree of interference between the feature vectors. The interference is either taken into consideration in whole, or not considered at all, which will inevitably lead to information loss. In our work, an entropy-weighted NAP (EWNAP) is proposed to deal with such “bipolar problem” in constructing the weighted matrix. The eigenvalues of covariance matrix of collected signals contain dynamical information, and the fuzzy entropy is adopted to evaluate the dispersion degree of these eigenvalues. After normalization, these entropy values are used to express the weight relationship in the weighted matrix of EWNAP. The features processed by EWNAP can be used as samples and combined with neural network to achieve fault diagnosis of rolling bearings. Furthermore, a fault diagnosis approach with insufficient data is demonstrated to validate the effectiveness of the proposed scheme. In the case studies, Case Western Reserve University bearing database and data collected from the bearing fault simulation bench are used. These case studies show that the proposed EWNAP alleviates the interference caused by various operating conditions, and the comparative analysis confirms that the proposed method works better than the conventional methods.

show abstract

“…In order to extract effective fault features, various signal processing methods have been proposed, such as spectral kurtosis, envelop spectrum, empirical mode decomposition, and so on [2][3][4][5][6]. Then, some conventional machine learning methods, for example, support vector machine(SVM), and hidden Markov model (HMM), have been adopted for intelligent bearing fault classification [7,8]. Although VBD is a well-developed and prevalent rolling bearing fault detection approach, it has some limitations in practical applications.…”

Section: Introductionmentioning

confidence: 99%

A Non-Contact Fault Diagnosis Method for Rolling Bearings Based on Acoustic Imaging and Convolutional Neural Networks

et al. 2020

View full text Add to dashboard Cite

Rolling bearing fault diagnosis is conventionally performed by vibration-based diagnosis (VBD). However, VBD is restrained in some cases because vibration measurement usually requires the contact with the machine. Acoustical-based fault diagnosis (ABD) has the advantage of non-contact measurement over VBD. However, ABD has received little attention and rarely applied in bearing fault diagnosis. In this paper, a new non-contact ABD method for rolling bearings using acoustic imaging and convolutional neural networks (CNN) is proposed. Firstly, a microphone array is used to acquire the acoustic field radiated by rolling bearings. Then, acoustic imaging is performed with the wave superposition method (WSM). The reconstructed acoustic images can depict the spatial distribution of the acoustic field, which add a new spatial dimension in the acoustic data representation for fault diagnosis and makes it possible to localize the sound sources. Finally, CNN is applied to accomplish bearing fault diagnosis, which can overcome the problems of handcrafted feature extraction in traditional ABD methods. Experimental results verify the effectiveness of the proposed ABD method. Comparisons with peer state-of-the-art ABD methods further validate that the proposed method can mitigate the drawbacks of the existing ABD methods, and obtain more accurate and reliable diagnosis results.INDEX TERMS Bearing fault diagnosis; acoustic imaging; CNN; wave superposition method; acousticalbased fault diagnosis

show abstract

A Robust Performance Degradation Modeling Approach Based on Student’s t-HMM and Nuisance Attribute Projection

Cited by 7 publications

References 39 publications

Bearing Performance Degradation Assessment Based on SC-RMI and Student’s t-HMM

Bearing Performance Degradation Assessment Based on SC-RMI and Student’s t-HMM

Robust fault diagnosis of rolling bearings via entropy-weighted nuisance attribute projection and neural network under various operating conditions

A Non-Contact Fault Diagnosis Method for Rolling Bearings Based on Acoustic Imaging and Convolutional Neural Networks

Contact Info

Product

Resources

About