scholarly journals A novel classification method combining adaptive local iterative filtering with singular value decomposition for fault diagnosis

2018 ◽  
Vol 20 (3) ◽  
pp. 1355-1369
Author(s):  
Cancan Yi ◽  
Yong Lv ◽  
Yi Zhang ◽  
Han Xiao ◽  
Zhang Dang
Keyword(s):  
Fault Diagnosis ◽  
Singular Value Decomposition ◽  
Singular Value ◽  
Classification Method ◽  
Iterative Filtering ◽  
Value Decomposition

Fault diagnosis of a wind turbine rolling bearing using adaptive local iterative filtering and singular value decomposition

2016 ◽  
Vol 39 (11) ◽  
pp. 1643-1648 ◽  
Author(s):  
Xueli An ◽  
Hongtao Zeng ◽  
Weiwei Yang ◽  
Xuemin An
Keyword(s):  
Fault Diagnosis ◽  
Singular Value Decomposition ◽  
Wind Turbine ◽  
Multiple Scales ◽  
Roller Bearing ◽  
Vibration Signal ◽  
Singular Value ◽  
Feature Vectors ◽  
Iterative Filtering ◽  
Value Decomposition

Adaptive local iterative filtering (ALIF) is a new signal decomposition method that uses the iterative filters strategy together with an adaptive and data-driven filter length selection to achieve the decomposition. The complexity of wind power generation systems means that the randomness and kinetic mutation behaviour of their vibration signals are demonstrated at different scales. Thus it is necessary to analyse the vibration signal across multiple scales. A method based on ALIF and singular value decomposition (SVD) was used for the fault diagnosis of a wind turbine roller bearing. The ALIF method is used to decompose the bearing vibration signal into several stable components. The components, which contain major fault information, are selected to build an initial feature vector matrix. The singular value of the matrix is computed as the feature vectors of each bearing fault. The feature vectors embody the characteristics of the vibration signal. The nearest neighbour algorithm is used as a classifier to identify faults in a roller bearing. Experimental data show that the proposed method can be used to identify roller bearing faults of a wind turbine.

scholarly journals Research on Rolling Bearing Fault Feature Extraction Based on Singular Value Decomposition considering the Singular Component Accumulative Effect and Teager Energy Operator

2019 ◽  
Vol 2019 ◽  
pp. 1-14
Author(s):  
Longlong Li ◽  
Yahui Cui ◽  
Runlin Chen ◽  
Lingping Chen ◽  
Lihua Wang
Keyword(s):  
Fault Diagnosis ◽  
Singular Value Decomposition ◽  
Energy Operator ◽  
Vibration Signal ◽  
Rolling Bearing ◽  
Singular Value ◽  
Teager Energy Operator ◽  
Singular Component ◽  
Value Decomposition ◽  
Teager Energy

The extraction of impulsive signatures from a vibration signal is vital for fault diagnosis of rolling element bearings, which are always whelmed by noise, especially in the early stage of defect development. Aiming at the weak defect diagnosis, kurtosis of Teager energy operator (KTEO) spectrum is employed to indicate the fault information capacity of a spectrum, and considering the accumulative effect of a singular component, accumulative kurtosis of TEO (AKTEO) is firstly proposed to determine the proper signal reconstructed order during vibration signal processing using singular value decomposition (SVD). Then, a vibration processing scheme named SVD-AKTEO is designed where an iteration is employed to reflect an accumulative singular effect by kurtosis of TEO spectrum. Finally, the fault diagnosis results can be extracted from the TEO spectrum output by SVD-AKTEO. Simulation data and real data from a run-to-failure experiment of a rolling bearing are adopted to validate the efficiency, and comparative analysis demonstrates the feasibility to detect the early defect of the rolling bearing.

Sign in / Sign up

Export Citation Format

Share Document