Applying improved multi-scale entropy and support vector machines for bearing health condition identification

2010 ◽  
Vol 224 (6) ◽  
pp. 1315-1325 ◽  
Author(s):  
L Zhang ◽  
G Xiong ◽  
H Liu ◽  
H Zou ◽  
W Guo
Keyword(s):  
Support Vector Machines ◽  
Mechanical Vibration ◽  
Health Condition ◽  
Small Sample ◽  
Support Vector ◽  
Engineering Practice ◽  
Vibration Signals ◽  
Multi Scale ◽  
Vector Machines ◽  
Machine Health

Considering the non-linearity existing in bearing vibration signals as well as the scarcity of fault samples, this paper presents a method for bearing health condition identification based on improved multi-scale entropy (IMSE) and support vector machines (SVMs). IMSE refers to the calculation of improved sample entropies (i.e. fuzzy sample entropies across a sequence of scales). Applying IMSE to mechanical vibration signals can take into account not only the non-linearity but also the interactions and coupling between mechanical components, thus providing much more information regarding the machine health condition than traditional single-scale entropy can be expected to. In engineering practice, the amount of fault samples is often limited for training a classifier, which thus decreases the performance of traditional classifiers like artificial neural networks (ANNs). SVMs are derived from statistical learning theory, which is different from the conventional statistical theory on which ANNs are based. SVMs provide a favourable solution to small sample-sized problems. In this study, IMSE and SVMs are employed as fault feature extractor and classifier, respectively. The experimental results verify that the proposed method has potential applications in bearing health condition identification.

Bearing Condition Identification Using Improved Multiscale Entropy and SVMs

2009 ◽  
Vol 419-420 ◽  
pp. 817-820 ◽  
Author(s):  
Long Zhang ◽  
Guo Liang Xiong ◽  
He Sheng Liu ◽  
Hui Jun Zou
Keyword(s):  
Health Condition ◽  
Small Sample ◽  
Multiscale Entropy ◽  
Support Vector ◽  
Engineering Practice ◽  
Single Scale ◽  
Vector Machines ◽  
Feature Extractor ◽  
Potential Applications ◽  
Machine Health

This paper presents a method for bearing health condition identification based on improved multiscale entropy (IMSE) and support vector machines (SVMs). IMSE refers to the calculation of improved sample entropies, i.e., fuzzy sample entropies (FSampEn) across a sequence of time scales, which takes into account not only the nonlinearity but also the interactions and coupling between mechanical components, thus providing much more information regarding machine health condition compared to traditional single scale-based entropy. On the other hand, in engineering practice, the amount of fault samples is often limited, which thus decrease the performance of traditional classifiers like artificial neural networks (ANNs). Currently popular SVMs provide a favorable solution to small sample-sized problems. In this study, IMSE and SVMs were employed as fault feature extractor and classifier, respectively. The experimental results verify that the proposed method has potential applications in bearing health condition identification.

A Study of the Screening Efficiency of a Probability Sieve Based on Higher-Order Spectrum Analysis and Support Vector Machines

2011 ◽  
Vol 291-294 ◽  
pp. 2089-2093
Author(s):  
Zheng Zhong Shi ◽  
Yi Jian Huang
Keyword(s):  
Support Vector Machines ◽  
Least Square Method ◽  
Higher Order ◽  
Least Square ◽  
Support Vector ◽  
Vibration Signals ◽  
Order Spectrum ◽  
Vector Machines ◽  
Screening Efficiency ◽  
Relative Errors

Aiming at drawbacks of current methods for predicting the screening efficiency of probability sieve, this paper proposed a method of predict and study the screening efficiency of probability sieve based on higher-order spectrum(HOS) analysis and support vector machines(SVMs). First setting up trispectrum model with the vibration signals, then fitting out polynomial with least square method using the data which get out by the reconstruct power spectrum. Finaly, using support vector machines to predicting the screening efficiency with the coefficient of the polynomial as the sample input. The results show that the relative errors are all less than 2.4% and the absolute errors are all less than 0.021, which is ideal for efficiency forecast.

scholarly journals Short-Circuited Turn Fault Diagnosis in Transformers by Using Vibration Signals, Statistical Time Features, and Support Vector Machines on FPGA

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3598
Author(s):  
Jose R. Huerta-Rosales ◽  
David Granados-Lieberman ◽  
Arturo Garcia-Perez ◽  
David Camarena-Martinez ◽  
Juan P. Amezquita-Sanchez ◽  
...  
Keyword(s):  
Support Vector Machines ◽  
Electrical Network ◽  
Support Vector ◽  
Fisher Score ◽  
Vibration Signals ◽  
Linear Discriminant ◽  
Vector Machines ◽  
Field Programmable ◽  
Statistical Time ◽  
High Level

One of the most critical devices in an electrical system is the transformer. It is continuously under different electrical and mechanical stresses that can produce failures in its components and other electrical network devices. The short-circuited turns (SCTs) are a common winding failure. This type of fault has been widely studied in literature employing the vibration signals produced in the transformer. Although promising results have been obtained, it is not a trivial task if different severity levels and a common high-level noise are considered. This paper presents a methodology based on statistical time features (STFs) and support vector machines (SVM) to diagnose a transformer under several SCTs conditions. As STFs, 19 indicators from the transformer vibration signals are computed; then, the most discriminant features are selected using the Fisher score analysis, and the linear discriminant analysis is used for dimension reduction. Finally, a support vector machine classifier is employed to carry out the diagnosis in an automatic way. Once the methodology has been developed, it is implemented on a field-programmable gate array (FPGA) to provide a system-on-a-chip solution. A modified transformer capable of emulating different SCTs severities is employed to validate and test the methodology and its FPGA implementation. Results demonstrate the effectiveness of the proposal for diagnosing the transformer condition as an accuracy of 96.82% is obtained.

Fault diagnosis based on support vector machines with parameter optimization by an ant colony algorithm

2009 ◽  
Vol 224 (1) ◽  
pp. 217-229 ◽  
Author(s):  
X L Zhang ◽  
X F Chen ◽  
Z J He
Keyword(s):  
Fault Diagnosis ◽  
Support Vector Machines ◽  
Parameter Optimization ◽  
Ant Colony Algorithm ◽  
Small Sample ◽  
Ant Colony ◽  
Support Vector ◽  
Optimal Parameters ◽  
Vector Machines ◽  
Diagnosis Method

Since support vector machines (SVM) exhibit a good generalization performance in the small sample cases, these have a wide application in machinery fault diagnosis. However, a problem arises from setting optimal parameters for SVM so as to obtain optimal diagnosis result. This article presents a fault diagnosis method based on SVM with parameter optimization by ant colony algorithm to attain a desirable fault diagnosis result, which is performed on the locomotive roller bearings to validate its feasibility and efficiency. The experiment finds that the proposed algorithm of ant colony optimization with SVM (ACO—SVM) can help one to obtain a good fault diagnosis result, which confirms the advantage of the proposed ACO—SVM approach.

Sign in / Sign up

Export Citation Format

Share Document