Author:
Zhang Hengxi,Li Shengtao,Wu Tinghui,An Yuxiang
Abstract
Abstract
A defect identification and classification method for 316 stainless steel welded pipes is proposed based on eddy current testing technology. Initially, this approach acquires eddy current signals from steel pipes, applies Empirical Mode Decomposition (EMD) to derive Intrinsic Mode Functions (IMFs), chooses the principal IMF based on interrelationships, and extracts time-frequency domain characteristic parameters from the selected IMF. To enhance model recognition efficiency, Principal Component Analysis (PCA) is employed to reduce the dimensionality of the feature vector set. Ultimately, a Support Vector Machine (SVM) is utilized to identify and classify weld defects. The results indicate that this method is highly accurate in identifying defects in 316 stainless steel welded pipes.
Reference9 articles.
1. Inversion Study of Weak Magnetic Signals of Artificial Defects in 304 Stainless Steel Based on Libsvm;Guo;Mechanical Strength,2021
2. Novel austenitic steel aging classification method using eddy current testing and a support vector machine;Arenas Mónica;Measurement,2018
3. Method for Surface Crack Detection of Welded Structure in Pulsed Eddy Current Thermography;Liu;China Mechanical Engineering,2016
4. Comparative study of eddy current and barkhausen noise nondestructive testing methods in microstructural examination of ferrite–martensite dual-phase steel;Ghanei;Journal of Magnetism & Magnetic Materials,2014
5. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series Analysis;Huang;Proceedings Mathematical Physical & Engineering Sciences,1998