Feature Selection for Fault Detection in Rolling Element Bearings Using Mutual Information

2011 ◽  
Vol 133 (6) ◽  
Author(s):  
Karthik Kappaganthu ◽  
C. Nataraj
Keyword(s):  
Fault Detection ◽  
Mutual Information ◽  
Quantitative Measure ◽  
Rolling Element Bearings ◽  
Outer Race ◽  
Time Frequency ◽  
Rolling Element ◽  
Degree Of Confidence ◽  
Hidden Layer ◽  
Optimal Feature

Rolling element bearings are among the key components in many rotating machineries. It is hence necessary to determine the condition of the bearing with a reasonable degree of confidence. Many techniques have been developed for bearing fault detection. Each of these techniques has its own strengths and weaknesses. In this paper, various features are compared for detecting inner and outer race defects in rolling element bearings. Mutual information between the feature and the defect is used as a quantitative measure of quality. Various time, frequency, and time-frequency domain features are compared and ranked according to their cumulative mutual information content, and an optimal feature set is determined for bearing classification. The performance of this optimal feature set is evaluated using an artificial neural network with one hidden layer. An overall classification accuracy of 97% was obtained over a range of rotating speeds.

Mutual Information Based Feature Selection From Data Driven and Model Based Techniques for Fault Detection in Rolling Element Bearings

2011 ◽  
Author(s):  
Karthik Kappaganthu ◽  
C. Nataraj
Keyword(s):  
Fault Detection ◽  
Mutual Information ◽  
Diagnostic System ◽  
Classification Performance ◽  
Data Driven ◽  
Time Frequency ◽  
Bearing Fault ◽  
Model Based ◽  
Rolling Element ◽  
And Performance

This paper proposes a novel technique combining datadriven and model-based techniques to significantly improve the performance in bearing fault diagnostics. Features that provide best classification performance for the given data are selected from a combined set of data driven and model based features. Some of the common data driven techniques from time, frequency and time-frequency domain are considered. For model based feature extraction, recently developed cross-sample entropy is used. The ranking and performance of each of these feature sets are studied, when used independently and when used together. Mutual information based technique is used for ranking and selection of the optimal feature set. Using this method, the contribution to performance and redundancy of each of the data driven features and model based features can be studied. This method can be used to design an effective diagnostic system for bearing fault detection.

Experimental investigation on time-domain features in the diagnosis of rolling element bearings by acoustic emission

2021 ◽  
pp. 107754632110161
Author(s):  
Aref Aasi ◽  
Ramtin Tabatabaei ◽  
Erfan Aasi ◽  
Seyed Mohammad Jafari
Keyword(s):  
Acoustic Emission ◽  
Condition Monitoring ◽  
Time Domain ◽  
Central Moment ◽  
Rolling Element Bearings ◽  
Test Rig ◽  
Acoustic Emission Sensor ◽  
Outer Race ◽  
Common Time ◽  
Rolling Element

Inspired by previous achievements, different time-domain features for diagnosis of rolling element bearings are investigated in this study. An experimental test rig is prepared for condition monitoring of angular contact bearing by using an acoustic emission sensor for this purpose. The acoustic emission signals are acquired from defective bearing, and the sensor takes signals from defects on the inner or outer race of the bearing. By studying the literature works, different domains of features are classified, and the most common time-domain features are selected for condition monitoring. The considered features are calculated for obtained signals with different loadings, speeds, and sizes of defects on the inner and outer race of the bearing. Our results indicate that the clearance, sixth central moment, impulse, kurtosis, and crest factors are appropriate features for diagnosis purposes. Moreover, our results show that the clearance factor for small defects and sixth central moment for large defects are promising for defect diagnosis on rolling element bearings.

Simultaneous Monitoring of Rolling-Element and Journal Bearings Using Analysis of Structure-Born Ultrasound Acoustic Emissions

2010 ◽  
Author(s):  
A. Albers ◽  
M. Dickerhof
Keyword(s):  
Acoustic Emission ◽  
Acoustic Emissions ◽  
Piezoelectric Sensor ◽  
Journal Bearings ◽  
Rolling Element Bearings ◽  
Acoustic Emission Sensor ◽  
Outer Race ◽  
Rolling Element ◽  
Characteristic Values ◽  
Acoustic Emission Technology

The application of Acoustic Emission technology for monitoring rolling element or hydrodynamic plain bearings has been addressed by several authors in former times. Most of these investigations took place under idealized conditions, to allow the concentration on one single source of emission, typically recorded by means of a piezoelectric sensor. This can be achieved by either eliminating other sources in advance or taking measures to shield them out (e. g. by placing the acoustic emission sensor very close to the source of interest), so that in consequence only one source of structure-born sound is present in the signal. With a practical orientation this is often not possible. In point of fact, a multitude of potential sources of emission can be worth considering, unfortunately superimposing one another. The investigations reported in this paper are therefore focused on the simultaneous monitoring of both bearing types mentioned above. Only one piezoelectric acoustic emission sensor is utilized, which is placed rather far away from the monitored bearings. By derivation of characteristic values from the sensor signal, different simulated defects can be detected reliably: seeded defects in the inner and outer race of rolling element bearings as well as the occurrence of mixed friction in the sliding surface bearing due to interrupted lubricant inflow.

Wigner–Ville distribution based on cyclic spectral density and the application in rolling element bearings diagnosis

2011 ◽  
Vol 225 (12) ◽  
pp. 2831-2847 ◽  
Author(s):  
Y Zhou ◽  
J Chen ◽  
G M Dong ◽  
W B Xiao ◽  
Z Y Wang
Keyword(s):  
Fault Diagnosis ◽  
Spectral Density ◽  
Wigner Distribution ◽  
Rolling Element Bearing ◽  
Rolling Element Bearings ◽  
Accurate Analysis ◽  
Time Frequency ◽  
Bearing Fault ◽  
Rolling Element ◽  
Cross Terms

The vibration signals of rolling element bearings are random cyclostationary when they have faults. Also, statistical properties of the signals change periodically with time. The accurate analysis of time-varying signals is an essential pre-requisite for the fault diagnosis and hence safe operation of rolling element bearings. The Wigner distribution is probably most widely used among the Cohen’s class in order to describe how the spectral content of a signal changes over time. However, the basic nature of such signals causes significant interfering cross-terms, which do not permit a straightforward interpretation of the energy distribution. To overcome this difficulty, the Wigner–Ville distribution (WVD) based on the cyclic spectral density (CSD) is discussed in this article. It is shown that the improved WVD, based on CSD of a long time series, can render the time–frequency distribution less susceptible to noise, and restrain the cross-terms in the time–frequency domain. Simulation and experiment of the rolling element-bearing fault diagnosis are performed, and the results indicate the validity of WVD based on CSD in time–frequency analysis for bearing fault detection.

Complementary ensemble local means decomposition method and its application to rolling element bearings fault diagnosis

2019 ◽  
Vol 233 (5) ◽  
pp. 868-880 ◽  
Author(s):  
Yao Cheng ◽  
Dong Zou
Keyword(s):  
Decomposition Method ◽  
Signal Reconstruction ◽  
Outer Race ◽  
Time Frequency ◽  
Frequency Decomposition ◽  
Local Means ◽  
Residual Noise ◽  
Rolling Element ◽  
Noisy Signals ◽  
Mode Mixing

Local means decomposition is an adaptive and nonparametric time–frequency decomposition method for nonstationary and nonlinear signals. However, in practice, local means decomposition is susceptible to mode mixing phenomena and produces different scale oscillations in one mode or similar scale oscillations in different modes, rendering the decomposition results difficult to interpret in terms of physical meansing. The noise-assisted ensemble local means decomposition method not only effectively resolved mode mixing but also generated a new problem, which tolerates residual noise in signal reconstruction. Targeting these shortcomings, this article proposes complementary ensemble local means decomposition, a novel noise-assisted time–frequency analysis method. First, an ensemble of white noise is added to the original signal via complementary positive and negative pairs. Second, local means decomposition is applied to decompose the noisy signals into a series of product functions, and the final results are obtained by averaging. The simulation results confirm that complementary ensemble local means decomposition offers an innovative improvement over ensemble local means decomposition in terms of eliminating residual noise. The superiority of the proposed method was further validated on fault signals obtained from faulty railway bearings (rolling element and outer race fault signals).

scholarly journals Estimation of the Defect Width on the Outer Race of a Rolling Element Bearing under Time-Varying Speed Conditions

2019 ◽  
Vol 2019 ◽  
pp. 1-11 ◽  
Author(s):  
Guang-Quan Hou ◽  
Chang-Myung Lee
Keyword(s):  
Research Work ◽  
Defect Size ◽  
Remaining Useful Life ◽  
Rolling Element Bearing ◽  
Time Varying ◽  
Entry And Exit ◽  
Outer Race ◽  
Time Frequency ◽  
Rolling Element ◽  
Element Bearing

Fault diagnosis and failure prognostics for rolling element bearing are helpful for preventing equipment failure and predicting the remaining useful life (RUL) to avoid catastrophic failure. Spall size is an important fault feature for RUL prediction, and most research work has focused on estimating the fault size under constant speed conditions. However, estimation of the defect width under time-varying speed conditions is still a challenge. In this paper, a method is proposed to solve this problem. To enhance the entry and exit events, the edited cepstrum is used to remove the determined components. The preprocessed signal is resampled from the time domain to the angular domain to eliminate the effect of speed variation and measure the defect size of a rolling element bearing on outer race. Next, the transient impulse components are extracted by local mean decomposition. The entry and exit points when the roller passes over the defect width on the outer race were identified by further processing the extracted signal with time-frequency analysis based on the continuous wavelet transform. The defect size can be calculated with the angle duration, which is measured from the identified entry and exit points. The proposed method was validated experimentally.

scholarly journals New State Identification Method for Rotating Machinery under Variable Load Conditions Based on Hybrid Entropy Features and Joint Distribution Adaptation

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-17
Author(s):  
Xiaoming Xue ◽  
Nan Zhang ◽  
Suqun Cao ◽  
Wei Jiang ◽  
Jianzhong Zhou ◽  
...  
Keyword(s):  
Joint Distribution ◽  
Operating Conditions ◽  
Variable Load ◽  
Rolling Element Bearings ◽  
Time Frequency ◽  
Identification Method ◽  
State Identification ◽  
Rolling Element ◽  
Load Conditions ◽  
Testing Set

Fault identification under variable operating conditions is a task of great importance and challenge for equipment health management. However, when dealing with this kind of issue, traditional fault diagnosis methods based on the assumption of the distribution coherence of the training and testing set are no longer applicable. In this paper, a novel state identification method integrated by time-frequency decomposition, multi-information entropies, and joint distribution adaptation is proposed for rolling element bearings. At first, fast ensemble empirical mode decomposition was employed to decompose the vibration signals into a collection of intrinsic mode functions, aiming at obtaining the multiscale description of the original signals. Then, hybrid entropy features that can characterize the dynamic and complexity of time series in the local space, global space, and frequency domain were extracted from each intrinsic mode function. As for the training and testing set under different load conditions, all data was mapped into a reproducing space by joint distribution adaptation to reduce the distribution discrepancies between datasets, where the pseudolabels of the testing set and the final diagnostic results were obtained by the k-nearest neighbor algorithm. Finally, five cases with the training and testing set under variable load conditions were used to demonstrate the performance of the proposed method, and comparisons with some other diagnosis models combined with the same features and other dimensionality reduction methods were also discussed. The analysis results show that the proposed method can effectively recognize the multifaults of rolling element bearings under variable load conditions with higher accuracies and has sound practicability.

Sign in / Sign up

Export Citation Format

Share Document