A New FeAture extrActioN Method For GeAr FAult diAGNosis ANd ProGNosis NowA MetodA diAGNozowANiA i ProGNozowANiA uszkodzeń PrzekłAdNi z wykorzystANieM ekstrAkcji cech

نویسندگان

  • Xinghui ZhAng
  • Jianshe KAng
  • eric Bechhoefer
  • Jianmin ZhAo
چکیده

Robust features are very critical to track the degradation process of a gear. They are key factors for implementing fault diagnosis and prognosis. This has driven the need in research for extracting good features. This paper used a new method, Narrowband Interference Cancellation, to suppress the narrow band component and enhance the impulsive component enabling the gear fault detection easier. This method can improve the signal to noise ratio of impulse train associated with the gear fault frequency. A run-to-failure test is used to demonstrate the method’s effectiveness. Based on the time synchronous signal of high speed shaft, Sideband Index is extracted from the signals processed by Narrowband Interference Cancellation. This feature has good degradation trend than traditional Sideband Index extracted from the time synchronous average signal directly. Comparison of features based on different extraction process proves the effectiveness of developed method.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Novel Traffic Prediction Scheme for Broadband Satellite Communications System

Multi-user traffic prediction is an important issue in broadband satellite communications system. In view of multiple UTs (User Terminal) which are treated by a single ST (Satellite Terminal), a novel NMF-based (Nonnegative Matrix Factorization-based) multi-user traffic prediction method is proposed. Compared with previous prediction schemes, the new method can reduce computational complexity a...

متن کامل

Pattern recognition for partial discharge in GIS based on pulse coupled neural networks and wavelet packet decomposition

Based on the characteristics of partial discharge (PD) defects in gas insulated switchgear (GIS), four typical single defects were designed for the present paper. PD three-dimensional (3D) patterns were constructed based on the ultra high frequency detection systems. The pulse-coupled neural networks (PCNN) and wavelet packet decomposition (WPD) method were used in PD feature extraction. The re...

متن کامل

A new method to diagnose rotor faults in 3-phase induction motors coupled to time-varying loads

This paper proposes a new method, based on the analysis of the active and reactive powers of the induction motor, for the diagnosis of rotor faults in motors coupled to time-varying loads. This new method is able to discriminate the presence of a true rotor fault from a time-varying load and can be used even when both the fault and the load torque variation coexist. A theoretical analysis compl...

متن کامل

An application of mathematical morphology to filtering and feature extraction for pattern recognition

Mathematical Morphology is a very efficient tool for image processing and analysis, based on non-linear local operators. In this paper mathematical morphology is applied to pattern recognition. The appropriate practical methods of morphological image processing are described, as well as its theoretical background. An example of an application of described methods to recognition of car number pl...

متن کامل

The Feature Selection based Power Quality Event Classification using Wavelet Transform and Logistic Model Tree

This paper presents a new power quality event classification technique using wavelet transform and logistic model tree. The proposed method uses the samples of three cycle duration of three line voltage of power quality events. The features of these samples are obtained by using the wavelet transform and a few different feature extraction techniques. The sequential forward selection method base...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2014