scholarly journals Multi-Domain Entropy-Random Forest Method for the Fusion Diagnosis of Inter-Shaft Bearing Faults with Acoustic Emission Signals

Entropy ◽  
2019 ◽  
Vol 22 (1) ◽  
pp. 57 ◽  
Author(s):  
Jing Tian ◽  
Lili Liu ◽  
Fengling Zhang ◽  
Yanting Ai ◽  
Rui Wang ◽  
...  
Keyword(s):  
Acoustic Emission ◽  
Fault Diagnosis ◽  
Random Forest ◽  
High Sensitivity ◽  
Generalization Ability ◽  
Bearing Faults ◽  
Bearing Fault Diagnosis ◽  
Diagnosis Method ◽  
Ae Signal ◽  
Ae Signals

Inter-shaft bearing as a key component of turbomachinery is a major source of catastrophic accidents. Due to the requirement of high sampling frequency and high sensitivity to impact signals, AE (Acoustic Emission) signals are widely applied to monitor and diagnose inter-shaft bearing faults. With respect to the nonstationary and nonlinear of inter-shaft bearing AE signals, this paper presents a novel fault diagnosis method of inter-shaft bearing called the multi-domain entropy-random forest (MDERF) method by fusing multi-domain entropy and random forest. Firstly, the simulation test of inter-shaft bearing faults is conducted to simulate the typical fault modes of inter-shaft bearing and collect the data of AE signals. Secondly, multi-domain entropy is proposed as a feature extraction approach to extract the four entropies of AE signal. Finally, the samples in the built set are divided into two subsets to train and establish the random forest model of bearing fault diagnosis, respectively. The effectiveness and generalization ability of the developed model are verified based on the other experimental data. The proposed fault diagnosis method is validated to hold good generalization ability and high diagnostic accuracy (~0.9375) without over-fitting phenomenon in the fault diagnosis of bearing shaft.

Fault Diagnosis Method for Inter-Shaft Bearings Based on Information Exergy and Random Forest

2018 ◽  
Author(s):  
Jing Tian ◽  
Yanting Ai ◽  
Ming Zhao ◽  
Chengwei Fei ◽  
Fengling Zhang
Keyword(s):  
Fault Diagnosis ◽  
Energy Spectrum ◽  
Random Forest ◽  
Complex Signals ◽  
Bearing Faults ◽  
Outer Race ◽  
Bearing Fault ◽  
Wavelet Energy ◽  
Diagnosis Method ◽  
Ae Signals

To reasonably process the complex signals and improve the diagnosis accuracy of inter-shaft bearing incipient faults, this paper develops wavelet energy spectrum exergy (WESE) and random forest (RF) (short for WESE-RF) method with respect to acoustic emission (AE) signals. Inter-shaft bearing faults, which contain inner race fault, outer race fault, rolling element faults and normal status under different measuring points and different rotational speeds, are simulated based on the test rig of inter-shaft bearings, to collect the AE signals of these faults. Regarding the AE signals of inter-shaft bearing faults, the WESE values, one signal feature, are extracted from an information exergy perspective, and are applied to structure feature vectors. The WESE values of these AE signals are regarded as the sample set which include the training samples subset used to establish the WESE-RF model of fault diagnosis and the test samples subset applied to test the effectiveness of the developed WESE-RF model. The investigation on the fault diagnosis of inter-shaft bearing demonstrates the fault diagnosis method with the WESE-RF has good generalization ability and high diagnostic accuracy of over 0.9 for inter-shaft bearing fault. The efforts of this paper provide a useful approach-based information exergy and wavelet energy spectrum for inter-shaft bearing fault diagnosis.

scholarly journals Intelligent Fault Diagnosis Method Using Acoustic Emission Signals for Bearings under Complex Working Conditions

2020 ◽  
Vol 10 (20) ◽  
pp. 7068
Author(s):  
Minh Tuan Pham ◽  
Jong-Myon Kim ◽  
Cheol Hong Kim
Keyword(s):  
Acoustic Emission ◽  
Feature Extraction ◽  
Deep Learning ◽  
Fault Diagnosis ◽  
Extraction Methods ◽  
High Accuracy ◽  
Time Frequency ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Ae Signals

Recent convolutional neural network (CNN) models in image processing can be used as feature-extraction methods to achieve high accuracy as well as automatic processing in bearing fault diagnosis. The combination of deep learning methods with appropriate signal representation techniques has proven its efficiency compared with traditional algorithms. Vital electrical machines require a strict monitoring system, and the accuracy of these machines’ monitoring systems takes precedence over any other factors. In this paper, we propose a new method for diagnosing bearing faults under variable shaft speeds using acoustic emission (AE) signals. Our proposed method predicts not only bearing fault types but also the degradation level of bearings. In the proposed technique, AE signals acquired from bearings are represented by spectrograms to obtain as much information as possible in the time–frequency domain. Feature extraction and classification processes are performed by deep learning using EfficientNet and a stochastic line-search optimizer. According to our various experiments, the proposed method can provide high accuracy and robustness under noisy environments compared with existing AE-based bearing fault diagnosis methods.

A heterogeneous fault diagnosis method for bearings in gearbox

2014 ◽  
Vol 229 (8) ◽  
pp. 1491-1499 ◽  
Author(s):  
Xiaohui Chen ◽  
Lei Xiao ◽  
Xinghui Zhang ◽  
Zhenxiang Liu
Keyword(s):  
Fault Diagnosis ◽  
Bearing Failure ◽  
Optimum Frequency ◽  
Bearing Faults ◽  
Outer Race ◽  
Bearing Fault ◽  
Spectral Kurtosis ◽  
Bearing Fault Diagnosis ◽  
Diagnosis Method ◽  
Inner Race

Bearing failure is one of the most important causes of breakdown of rotating machinery. These failures can lead to catastrophic disasters or result in costly downtime. One of the key problems in bearing fault diagnosis is to detect the bearing fault as early as possible. This capability enables the operator to have enough time to do some preventive maintenance. Most papers investigate the bearing faults under rational assumption that bearings work individually. However, bearings are usually working as a part of complex systems like a gearbox. The fault signal of bearings can be easily masked by other vibration generated from gears and shafts. The proposed method separates bearing signals from other signals, and then the optimum frequency band which the bearing fault signal is prominent is determined by mean envelope Kurtosis. Subsequently, the envelope analysis is used to detect the bearing faults. Finally, two bearing fault experiments are used to validate the proposed method. Each experiment contains two bearing fault modes, inner race fault and outer race fault. The results demonstrate that the proposed method can detect the bearing fault easier than spectral Kurtosis and envelope Kurtosis.

scholarly journals Distance and Density Similarity Based Enhancedk-NN Classifier for Improving Fault Diagnosis Performance of Bearings

2016 ◽  
Vol 2016 ◽  
pp. 1-11 ◽  
Author(s):  
Sharif Uddin ◽  
Md. Rashedul Islam ◽  
Sheraz Ali Khan ◽  
Jaeyoung Kim ◽  
Jong-Myon Kim ◽  
...  
Keyword(s):  
Acoustic Emission ◽  
Fault Diagnosis ◽  
Similarity Measure ◽  
Classification Accuracy ◽  
Diagnostic Performance ◽  
Nearest Neighbor ◽  
Neighborhood Size ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Ae Signals

An enhancedk-nearest neighbor (k-NN) classification algorithm is presented, which uses a density based similarity measure in addition to a distance based similarity measure to improve the diagnostic performance in bearing fault diagnosis. Due to its use of distance based similarity measure alone, the classification accuracy of traditionalk-NN deteriorates in case of overlapping samples and outliers and is highly susceptible to the neighborhood size,k. This study addresses these limitations by proposing the use of both distance and density based measures of similarity between training and test samples. The proposedk-NN classifier is used to enhance the diagnostic performance of a bearing fault diagnosis scheme, which classifies different fault conditions based upon hybrid feature vectors extracted from acoustic emission (AE) signals. Experimental results demonstrate that the proposed scheme, which uses the enhancedk-NN classifier, yields better diagnostic performance and is more robust to variations in the neighborhood size,k.

scholarly journals Rolling-Element Bearing Fault Diagnosis Using Improved LeNet-5 Network

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1693 ◽  
Author(s):  
Lanjun Wan ◽  
Yiwei Chen ◽  
Hongyang Li ◽  
Changyun Li
Keyword(s):  
Fault Diagnosis ◽  
Convergence Speed ◽  
Rolling Element Bearing ◽  
Fault Classification ◽  
Generalization Ability ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Rolling Element ◽  
Diagnosis Method ◽  
Element Bearing

To address the problems of low recognition accuracy, slow convergence speed and weak generalization ability of traditional LeNet-5 network used in rolling-element bearing fault diagnosis, a rolling-element bearing fault diagnosis method using improved 2D LeNet-5 network is put forward. The following improvements to the traditional LeNet-5 network are made: the convolution and pooling layers are reasonably designed and the size and number of convolution kernels are carefully adjusted to improve fault classification capability; the batch normalization (BN) is adopted after each convolution layer to improve convergence speed; the dropout operation is performed after each full-connection layer except the last layer to enhance generalization ability. To further improve the efficiency and effectiveness of fault diagnosis, on the basis of improved 2D LeNet-5 network, an end-to-end rolling-element bearing fault diagnosis method based on the improved 1D LeNet-5 network is proposed, which can directly perform 1D convolution and pooling operations on raw vibration signals without any preprocessing. The results show that the improved 2D LeNet-5 network and improved 1D LeNet-5 network achieve a significant performance improvement than traditional LeNet-5 network, the improved 1D LeNet-5 network provides a higher fault diagnosis accuracy with a less training time in most cases, and the improved 2D LeNet-5 network performs better than improved 1D LeNet-5 network under small training samples and strong noise environment.

Multi-feature entropy distance approach with vibration and acoustic emission signals for process feature recognition of rolling element bearing faults

2017 ◽  
Vol 17 (2) ◽  
pp. 156-168 ◽  
Author(s):  
Cheng-Wei Fei ◽  
Yat-Sze Choy ◽  
Guang-Chen Bai ◽  
Wen-Zhong Tang
Keyword(s):  
Acoustic Emission ◽  
Fault Diagnosis ◽  
Frequency Domain ◽  
Rolling Bearing ◽  
Distance Method ◽  
Bearing Faults ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Rolling Element ◽  
Entropy Distance

To accurately reveal rolling bearing operating status, multi-feature entropy distance method was proposed for the process character analysis and diagnosis of rolling bearing faults by the integration of four information entropies in time domain, frequency domain and time–frequency domain and two kinds of signals including vibration signals and acoustic emission signals. The multi-feature entropy distance method was investigated and the basic thought of rolling bearing fault diagnosis with multi-feature entropy distance method was given. Through rotor simulation test rig, the vibration and acoustic emission signals of six rolling bearing faults (ball fault, inner race fault, outer race fault, inner ball faults, inner–outer faults and normal) are gained under different rotational speeds. In the view of the multi-feature entropy distance method, the process diagnosis of rolling bearing faults was implemented. The analytical results show that multi-feature entropy distance fully reflects the process feature of rolling bearing faults with the change of rotating speed; the multi-feature entropy distance with vibration and acoustic emission signals better reports signal features than single type of signal (vibration or acoustic emission signal) in rolling bearing fault diagnosis; the proposed multi-feature entropy distance method holds high diagnostic precision and strong robustness (anti-noise capacity). This study provides a novel and useful methodology for the process feature extraction and fault diagnosis of rolling element bearings and other rotating machinery.

On the Use of Spectral Averaging of Acoustic Emission Signals for Bearing Fault Diagnostics

2014 ◽  
Vol 136 (6) ◽  
Author(s):  
Brandon Van Hecke ◽  
David He ◽  
Yongzhi Qu
Keyword(s):  
Acoustic Emission ◽  
Fault Diagnosis ◽  
Reduction Technique ◽  
Signal Frequency ◽  
Fault Diagnostics ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Area Of Interest ◽  
Ae Signals ◽  
Spectral Averaging

For years, vibration analysis has been the industry standard for bearing fault diagnosis. However, due to the various advantages over vibration based techniques, the quantification of acoustic emission (AE) for bearing health diagnosis has been an area of interest for recent years. Additionally, most AE based methodologies to date utilize data mining technologies. Presented in this paper is a new approach, combining a heterodyne based frequency reduction technique, time synchronous resampling, and spectral averaging to process AE signals and compute condition indicators (CIs) for bearing fault diagnostics. First, the heterodyne based frequency reduction technique allows the AE signal frequency to be down shifted from several MHz to less than 50 kHz, which approaches that of vibration based methodologies. Next, the sampled AE signals are band pass filtered to retain the useful information related to the bearing defects. Last, a trigger signal is utilized to time synchronously resample the AE signals to allow the calculation of a spectral average and the extraction and evaluation of CIs for bearing fault diagnosis. The technique presented in this paper is validated using the AE signals of seeded fault steel bearings on a bearing test rig. Presented is an effective AE based approach validated to diagnose all four fault types: inner race, outer race, ball, and cage. Moreover, the effectiveness of the presented approach is established through the comparison of both AE and vibration data.

scholarly journals Bearing Fault Diagnosis Method Based on Deep Convolutional Neural Network and Random Forest Ensemble Learning

Sensors ◽  
2019 ◽  
Vol 19 (5) ◽  
pp. 1088 ◽  
Author(s):  
Gaowei Xu ◽  
Min Liu ◽  
Zhuofu Jiang ◽  
Dirk Söffker ◽  
Weiming Shen
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Fault Diagnosis ◽  
Random Forest ◽  
Convolutional Neural Network ◽  
Ensemble Learning ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Diagnosis Method ◽  
Multi Level

Recently, research on data-driven bearing fault diagnosis methods has attracted increasing attention due to the availability of massive condition monitoring data. However, most existing methods still have difficulties in learning representative features from the raw data. In addition, they assume that the feature distribution of training data in source domain is the same as that of testing data in target domain, which is invalid in many real-world bearing fault diagnosis problems. Since deep learning has the automatic feature extraction ability and ensemble learning can improve the accuracy and generalization performance of classifiers, this paper proposes a novel bearing fault diagnosis method based on deep convolutional neural network (CNN) and random forest (RF) ensemble learning. Firstly, time domain vibration signals are converted into two dimensional (2D) gray-scale images containing abundant fault information by continuous wavelet transform (CWT). Secondly, a CNN model based on LeNet-5 is built to automatically extract multi-level features that are sensitive to the detection of faults from the images. Finally, the multi-level features containing both local and global information are utilized to diagnose bearing faults by the ensemble of multiple RF classifiers. In particular, low-level features containing local characteristics and accurate details in the hidden layers are combined to improve the diagnostic performance. The effectiveness of the proposed method is validated by two sets of bearing data collected from reliance electric motor and rolling mill, respectively. The experimental results indicate that the proposed method achieves high accuracy in bearing fault diagnosis under complex operational conditions and is superior to traditional methods and standard deep learning methods.

The Method of Fault Feature Extraction from Acoustic Emission Signals Using Wigner-Ville Distribution

2011 ◽  
Vol 216 ◽  
pp. 732-737 ◽  
Author(s):  
G.F. Bin ◽  
C.J. Liao ◽  
Xue Jun Li
Keyword(s):  
Acoustic Emission ◽  
Fault Diagnosis ◽  
High Energy ◽  
Frequency Resolution ◽  
Time Frequency ◽  
Fault Features ◽  
Very High Energy ◽  
Ae Signal ◽  
Experimental Data Analysis ◽  
Ae Signals

Wigner-Ville distribution (WVD) has the characteristics of very high-energy accumulation and excellent time-frequency resolution. It is a good way to extract fault feature of acoustic emission (AE) signals due to mechanical component broken. The characteristics of typical AE signals initiated by damages are analyzed. Based on the extracting principle of AE signals from damaged components, the WVD analysis method of AE signal is developed. WVD method is employed to the fault diagnosis of rolling bearings with AE technique. The fault features reading from experimental data analysis are clear, accurate and intuitionistic, meantime, the validity and accuracy of WVD method proposed are nice from the experimental results. Therefore, WVD method is useful for condition monitoring and fault diagnosis in conjunction with AE technique.

Sign in / Sign up

Export Citation Format

Share Document