Skip to main content
padlock icon - secure page this page is secure

Recognition of broken wire rope based on adaptive parameterless empirical wavelet transform and random forest

Buy Article:

$22.00 + tax (Refund Policy)

Magnetic flux leakage (MFL) detection is one of the most effective methods for non-destructive testing (NDT) of wire ropes. In view of the disadvantages of a magnetic flux leakage signal having a low signal-to-noise ratio (SNR) and the poor performance of the filtering algorithm, the use of an adaptive parameterless empirical wavelet transform (APEWT) and wavelet filtering algorithm is proposed to denoise the leakage magnetic field signal collected on the surface of the wire rope. The improved local modulus maxima method is used to locate and segment the magnetic field image of broken wires and the random forest (RF) classifier is designed to realise the quantitative identification of broken wires. Five texture features and four invariant moments totalled nine features of the magnetic flux leakage image that were trained as the input of the RF classifier and good results were obtained. Compared with the most commonly used backpropagation (BP) neural network, the feasibility and effectiveness of the RF classifier in the identification of broken wires of steel wire rope is demonstrated. The experimental results show that the filtering algorithm can effectively remove noise and the RF classifier that was designed has good recognition of this type of defect.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
No Metrics

Keywords: ADAPTIVE PARAMETERLESS EMPIRICAL WAVELET TRANSFORM; QUANTITATIVE IDENTIFICATION; RANDOM FOREST CLASSIFIER; WIRE ROPE

Document Type: Research Article

Publication date: July 1, 2019

More about this publication?
  • Access Key
  • Free content
  • Partial Free content
  • New content
  • Open access content
  • Partial Open access content
  • Subscribed content
  • Partial Subscribed content
  • Free trial content
Cookie Policy
X
Cookie Policy
Ingenta Connect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more