scholarly journals Tensor Transfer Learning for Intelligence Fault Diagnosis of Bearing with Semisupervised Partial Label Learning

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Chaofan Hu ◽  
Zhichao Zhou ◽  
Biao Wang ◽  
WeiGuang Zheng ◽  
Shuilong He
Keyword(s):  
Fault Diagnosis ◽  
Domain Adaptation ◽  
Subspace Learning ◽  
Rotating Machinery ◽  
Training Data ◽  
Tensor Representation ◽  
Vibration Signals ◽  
Tensor Representations ◽  
Partial Label Learning ◽  
And Training

A new tensor transfer approach is proposed for rotating machinery intelligent fault diagnosis with semisupervised partial label learning in this paper. Firstly, the vibration signals are constructed as a three-way tensor via trial, condition, and channel. Secondly, for adapting the source and target domains tensor representations directly, without vectorization, the domain adaptation (DA) approach named tensor-aligned invariant subspace learning (TAISL) is first proposed for tensor representation when testing and training data are drawn from different distribution. Then, semisupervised partial label learning (SSPLL) is first introduced for tackling a problem that it is hard to label a large number of instances and there exists much data left to be unlabeled. Ultimately, the proposed method is used to identify faults. The effectiveness and feasibility of the proposed method has been thoroughly validated by transfer fault experiments. The experimental results show that the presented technique can achieve better performance.

scholarly journals Cross-Machine Fault Diagnosis with Semi-Supervised Discriminative Adversarial Domain Adaptation

Sensors ◽  
2020 ◽  
Vol 20 (13) ◽  
pp. 3753
Author(s):  
Xiaodong Wang ◽  
Feng Liu ◽  
Dongdong Zhao
Keyword(s):  
Fault Diagnosis ◽  
Working Conditions ◽  
Domain Adaptation ◽  
Rotating Machinery ◽  
Working Environment ◽  
Training Data ◽  
Target Domain ◽  
Machine Fault Diagnosis ◽  
Machine Fault ◽  
The Cross

Bearings are ubiquitous in rotating machinery and bearings in good working conditions are essential for the availability and safety of the machine. Various intelligent fault diagnosis models have been widely studied aiming to prevent system failures. These data-driven fault diagnosis models work well when training data and testing data are from the same distribution, which is not easy to sustain in industry since the working environment of rotating machinery is often subject to change. Recently, the domain adaptation methods for fault diagnosis between different working conditions have been extensively researched, which fully utilize the labeled data from the same machine under different working conditions to address this domain shift diploma. However, for a target machine with seldom occurred faulty data under any working conditions, the domain adaptation approaches between working conditions are not applicable. Hence, the cross-machine fault diagnosis tasks are recently proposed to utilize the labeled data from related but not identical machines. The larger domain shift between machines makes the cross-machine fault diagnosis a more challenging task. The large domain shift may cause the well-trained model on source domain deteriorates on target domain, and the ambiguous samples near the decision boundary are prone to be misclassified. In addition, the sparse faulty samples in target domain make a class-imbalanced scenario. To address the two issues, in this paper we propose a semi-supervised adversarial domain adaptation approach for cross-machine fault diagnosis which incorporates the virtual adversarial training and batch nuclear-norm maximization to make the fault diagnosis robust and discriminative. Experiments of transferring between three bearing datasets show that the proposed method is able to effectively learn a discriminative model given only a labeled faulty sample of each class in target domain. The research provides a feasible approach for knowledge transfer in fault diagnosis scenarios.

An intelligent fault diagnosis method based on domain adaptation for rolling bearings under variable load conditions

2021 ◽  
pp. 095440622110329
Author(s):  
Jianqun Zhang ◽  
Qing Zhang ◽  
Xianrong Qin ◽  
Yuantao Sun
Keyword(s):  
Feature Extraction ◽  
Fault Diagnosis ◽  
Domain Adaptation ◽  
Rolling Bearing ◽  
Training Data ◽  
Variable Load ◽  
K Nearest Neighbor ◽  
Target Domain ◽  
Bearing Faults ◽  
Load Conditions

To identify rolling bearing faults under variable load conditions, a method named DISA-KNN is proposed in this paper, which is based on the strategy of feature extraction-domain adaptation-classification. To be specific, the time-domain and frequency-domain indicators are used for feature extraction. Discriminative and domain invariant subspace alignment (DISA) is used to minimize the data distributions’ discrepancies between the training data (source domain) and testing data (target domain). K-nearest neighbor (KNN) is applied to identify rolling bearing faults. DISA-KNN’s validation is proved by the experimental signal collected under different load conditions. The identification accuracies obtained by the DISA-KNN method are more than 90% on four datasets, including one dataset with 99.5% accuracy. The strength of the proposed method is further highlighted by comparisons with the other 8 methods. These results reveal that the proposed method is promising for the rolling bearing fault diagnosis in real rotating machinery.

Fault Detection and Discrimination of Rotating Machinery Using Frequency Symptom Parameter and Bayesian Network

2013 ◽  
Vol 470 ◽  
pp. 683-688
Author(s):  
Hai Yang Jiang ◽  
Hua Qing Wang ◽  
Peng Chen
Keyword(s):  
Fault Diagnosis ◽  
Fault Detection ◽  
Bayesian Network ◽  
Frequency Domain ◽  
Evaluation Method ◽  
Roller Bearing ◽  
Rotating Machinery ◽  
Discrimination Index ◽  
Vibration Signals ◽  
Diagnosis Method

This paper proposes a novel fault diagnosis method for rotating machinery based on symptom parameters and Bayesian Network. Non-dimensional symptom parameters in frequency domain calculated from vibration signals are defined for reflecting the features of vibration signals. In addition, sensitive evaluation method for selecting good non-dimensional symptom parameters using the method of discrimination index is also proposed for detecting and distinguishing faults in rotating machinery. Finally, the application example of diagnosis for a roller bearing by Bayesian Network is given. Diagnosis results show the methods proposed in this paper are effective.

Identification of Fault Signals in Rotating Machinery Using Higher Order Time Frequency Analysis

1997 ◽  
Author(s):  
Sang-Kwon Lee ◽  
Paul R. White
Keyword(s):  
Fault Diagnosis ◽  
Wigner Distribution ◽  
Higher Order ◽  
Rotating Machinery ◽  
Data Sets ◽  
Higher Order Statistics ◽  
Vibration Signals ◽  
Time Frequency ◽  
Increasing Trend ◽  
Monitoring Application

Abstract Impulsive acoustic and vibration signals within rotating machinery are often induced by irregular impacting. Thus the detection of these impulses can be useful for fault diagnosis. Recently there is an increasing trend towards the use of higher order statistics for fault detection within mechanical systems based on the observation that impulsive signals tend to increase the kurtosis values. We show that the fourth order Wigner Moment Spectrum, called the Wigner Trispectrum, has superior detection performance to second order Wigner distribution for typical impulsive signals found in a condition monitoring application. These methods are also applied to data sets measured within a car engine and industrial gearbox.

Joint Wavelet Transform and Time Synchronous Averaging Source Separation Method and its Application on Aero-Engine

2020 ◽  
Author(s):  
Jiqing Cong ◽  
Jianping Jing ◽  
Changmin Chen ◽  
Zezeng Dai ◽  
Jianhua Cheng
Keyword(s):  
Wavelet Transform ◽  
Fault Diagnosis ◽  
Fundamental Frequency ◽  
Averaging Method ◽  
Gaussian White Noise ◽  
Rotating Machinery ◽  
Vibration Source ◽  
Vibration Signals ◽  
Aero Engine ◽  
Harmonic Components

Abstract The reliability and safety of aero-engine are often the decisive factors for the safe and reliable flight of commercial aircraft. Hence, the vibration source location and fault diagnosis of aero-engine are of prime importance to detect faults and carry out fast and effective maintenance in time. However, the vibration signals collected by the sensors arranged on the casing of the aero-engine are generally the mixed signals of the main vibration sources inside the engine, and the components are extremely complicated. Therefore, the vibration source identification is a big challenge for a fault diagnosis and health management of the engine. In order to separate the key vibration sources of rotating machinery such as aero-engine, a Joint Wavelet Transform and Time Synchronous Averaging based algorithm (JWTS) is proposed in this paper. Based on the fact that the fundamental frequency and its harmonic and sub-harmonic components are generally included in the vibration spectrum of shaft fault signal of rotating machinery, wavelet transform and time synchronous averaging algorithm are combined to extract them. The algorithm completes separating the main vibration sources with three steps. First, the source number and fundamental frequency of each source are estimated using the wavelet transform. Second, every source is extracted from each observed signal by the time synchronous averaging method. Time synchronous averaging method can effectively extract a signal of cycle and harmonic rotor components and can suppress noise. Third, the optimal estimation of each source is determined according to signal’s 2-norm. Since the extracted source with a larger energy is closer to the real source, and signal’s 2-norm is a good indicator of the signal energy. Hence, the key vibration sources related to rotary speeds of the engine are obtained separately. The method is verified by synthetic mixed signals first. Three periodic signals of different frequencies are used to simulate the vibration sources of the aeroengine. The fundamental, harmonic and sub-harmonic components of them, as well as Gaussian white noise, are randomly mixed. The results show that the JWTS algorithm can estimate the number of the main sources and can extract each source effectively. Then the method is demonstrated using vibration signals of a real aero-engine. The results indicate that the proposed JWTS method has extracted and located the main sources within the aero-engine, including sources from the low-pressure rotor, high-pressure rotor, combustion chamber and accessory. Therefore, the proposed method provides a new fault diagnosis technology for rotating machinery, especially for a real aero-engine.

scholarly journals Fault Diagnosis of Rotating Machinery Based on One-Dimensional Deep Residual Shrinkage Network with a Wide Convolution Layer

2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Jingli Yang ◽  
Tianyu Gao ◽  
Shouda Jiang ◽  
Shijie Li ◽  
Qing Tang
Keyword(s):  
Fault Diagnosis ◽  
Noise Immunity ◽  
Rotating Machinery ◽  
Experimental Results ◽  
Vibration Signals ◽  
One Dimensional ◽  
Noise Interference ◽  
Diagnosis Method ◽  
Diagnosis Model ◽  
Short Time

In actual engineering applications, inevitable noise seriously affects the accuracy of fault diagnosis for rotating machinery. To effectively identify the fault classes of rotating machinery under noise interference, an efficient fault diagnosis method without additional denoising procedures is proposed. First, a one-dimensional deep residual shrinkage network, which directly takes the raw vibration signals contaminated by noise as input, is developed to realize end-to-end fault diagnosis. Then, to further enhance the noise immunity of the diagnosis model, the first layer of the model is set to a wide convolution layer to extract short time features. Moreover, an adaptive batch normalization algorithm (AdaBN) is introduced into the diagnosis model to enhance the adaptability to noise. Experimental results illustrate that the fault diagnosis model for rotating machinery based on one-dimensional deep residual shrinkage network with a wide convolution layer (1D-WDRSN) can accurately identify the fault classes even under noise interference.

scholarly journals A New Transferable Fault Diagnosis Approach of Rotating Machinery Based on Deep Autoencoder and Dominant Features Selection under Different Operating Conditions

2021 ◽  
Vol 2021 ◽  
pp. 1-21
Author(s):  
Fei Dong ◽  
Xiao Yu ◽  
Xinguo Shi ◽  
Ke Liu ◽  
Zhaoli Wu ◽  
...  
Keyword(s):  
Fault Diagnosis ◽  
Source Model ◽  
Rotating Machinery ◽  
Operating Conditions ◽  
Training Data ◽  
Fine Tuning ◽  
Discrete Wavelet ◽  
Features Selection ◽  
Target Model ◽  
Diagnosis Approach

In the actual industrial scenarios, most existing fault diagnosis approaches are faced with two challenges, insufficient labeled training data and distribution divergences between training and testing datasets. For the above issues, a new transferable fault diagnosis approach of rotating machinery based on deep autoencoder and dominant features selection is proposed in this article. First, maximal overlap discrete wavelet packet transform is applied for signals processing and mix-domains statistical feature extraction. Second, dominant features selection by importance score and differences between domains is proposed to select dominant features with high fault-discriminative ability and domain invariance. Then, selected dominant features are used for pretraining deep autoencoder (source model), which helps in enhancing the fault representative ability of deep features. The parameters of the source model are transferred to the target model, and normal state features from target domain are adopted for fine-tuning the target model. Finally, the target model is applied for fault patterns classification. Motor and bearing fault datasets are used for a series of experiments, and the results verify that the proposed methods have better cross-domain diagnosis performance than comparative models.

A domain adaptation model for early gear pitting fault diagnosis based on deep transfer learning network

2019 ◽  
Vol 234 (1) ◽  
pp. 168-182 ◽  
Author(s):  
Jialin Li ◽  
Xueyi Li ◽  
David He ◽  
Yongzhi Qu
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Transfer Learning ◽  
Working Conditions ◽  
Deep Neural Network ◽  
Domain Adaptation ◽  
Diagnostic Model ◽  
Target Domain ◽  
Vibration Signals ◽  
Learning Network

In recent years, research on gear pitting fault diagnosis has been conducted. Most of the research has focused on feature extraction and feature selection process, and diagnostic models are only suitable for one working condition. To diagnose early gear pitting faults under multiple working conditions, this article proposes to develop a domain adaptation diagnostic model–based improved deep neural network and transfer learning with raw vibration signals. A particle swarm optimization algorithm and L2 regularization are used to optimize the improved deep neural network to improve the stability and accuracy of the diagnosis. When using the domain adaptation diagnostic model for fault diagnosis, it is necessary to discriminate whether the target domain (test data) is the same as the source domain (training data). If the target domain and the source domain are consistent, the trained improved deep neural network can be used directly for diagnosis. Otherwise, the transfer learning is combined with improved deep neural network to develop a deep transfer learning network to improve the domain adaptability of the diagnostic model. Vibration signals for seven gear types with early pitting faults under 25 working conditions collected from a gear test rig are used to validate the proposed method. It is confirmed by the validation results that the developed domain adaptation diagnostic model has a significant improvement in the adaptability of multiple working conditions.

A novel gearbox fault feature extraction and classification using Hilbert empirical wavelet transform, singular value decomposition, and SOM neural network

2017 ◽  
Vol 24 (12) ◽  
pp. 2512-2531 ◽  
Author(s):  
Boualem Merainani ◽  
Chemseddine Rahmoune ◽  
Djamel Benazzouz ◽  
Belkacem Ould-Bouamama
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Wavelet Transform ◽  
Fault Diagnosis ◽  
Rotating Machinery ◽  
Singular Value ◽  
Vibration Signals ◽  
Empirical Wavelet Transform ◽  
Som Neural Network ◽  
Value Decomposition

There are growing demands for condition monitoring and fault diagnosis of rotating machinery to lower unscheduled breakdown. Gearboxes are one of the fundamental components of rotating machinery; their faults identification and classification always draw a lot of attention. However, non-stationary vibration signals and low energy of weak faults makes this task challenging in many cases. Thus, a new fault diagnosis method which combines the Hilbert empirical wavelet transform (HEWT), singular value decomposition (SVD), and self-organizing feature map (SOM) neural network is proposed in this paper. HEWT, a new self-adaptive time-frequency analysis was applied to the vibration signals to obtain the instantaneous amplitude matrices. Then, the singular value vectors, as the fault feature vectors were acquired by applying the SVD. Last, the SOM was used for automatic gearbox fault identification and classification. An electromechanical model comprising an induction motor coupled with a single stage spur gearbox is considered where the vibration signals of four typical operation modes were simulated. The conditions include the healthy gearbox, input shaft slant crack, tooth cracking, and tooth surface pitting. Obtained results show that the proposed method effectively identifies the gearbox faults at an early stage and realizes automatic fault diagnosis. Moreover, performance evaluation and comparison between the proposed HEWT–SVD method and Hilbert–Huang transform (HHT)–SVD approach show that the HEWT–SVD is better for feature extraction.

Fault Classification Model of Rotor Based on Support Vector Machine

2011 ◽  
Vol 66-68 ◽  
pp. 1982-1987
Author(s):  
Wei Niu ◽  
Guo Qing Wang ◽  
Zheng Jun Zhai ◽  
Juan Cheng
Keyword(s):  
Support Vector Machine ◽  
Fault Diagnosis ◽  
Rotating Machinery ◽  
Classification Model ◽  
Fault Classification ◽  
Support Vector ◽  
Vibration Signals ◽  
Good Classification ◽  
Speed Change ◽  
Diagnosis Model

The vibration signals of rotating machinery in operation consist of plenty of information about its running condition, and extraction and identification of fault signals in the process of speed change are necessary for the fault diagnosis of rotating machinery. This paper improves DDAG classification method and proposes a new fault diagnosis model based on support vector machine to solve the problem of restricting the rotating machinery fault intelligent diagnosis due to the lack of fault data samples. The testing results demonstrate that the model has good classification precision and can correctly diagnose faults.

Sign in / Sign up

Export Citation Format

Share Document