Performance reliability evaluation of the feature vector in one-dimensional components based on the grey system theory

According to the selection and evaluation of low strain stress wave signal feature for one-dimensional components damage detection, define the quantitative information entropy based on probability theory and mathematical statistical theory and information entropy, build the feature vector of stress...

Full description

Saved in:
Bibliographic Details
Published in2015 12th IEEE International Conference on Electronic Measurement & Instruments (ICEMI) Vol. 1; pp. 207 - 211
Main Authors Kang Weixin, Li Jingde, Liu Yumei, Wagan, Raja Asif
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2015
Subjects
evaluation
Feature extraction
Information entropy
Reliability
Strain
Stress
the grey system theory
the quantitative information entropy
time-window
Wavelet packets
Online AccessGet full text

Cover

More Information
Summary:According to the selection and evaluation of low strain stress wave signal feature for one-dimensional components damage detection, define the quantitative information entropy based on probability theory and mathematical statistical theory and information entropy, build the feature vector of stress wave signal with the quantitative information entropy based on the time-window, by using the multi-parameter correlation analysis method of the grey system theory, raise the evaluation model of feature vector. This method is tested using stress wave of pile detection, the experimental results show that the feature vector of the quantitative information entropy has advantages in synthetic performance. The evaluation method raised in this paper can be widely applied to detect damage and defects of one-dimensional components like pile foundations, beam structures and mechanical rotators.
DOI:10.1109/ICEMI.2015.7494254