Известия Томского политехнического университета: Инжиниринг георесурсов (Nov 2022)
EMPIRICAL METHOD FOR RECOGNIZING A GROUP OF STRESS-CORROSION CRACKS IN METAL BY THE SIGNAL OF EDDY-CURRENT SENSOR
Abstract
The relevance of the research is related to improving the safety of operation of steel gas pipelines containing stress-corrosion cracks and corrosion damage to metal detected during technical diagnostics by eddy-current non-destructive testing. The main aim of the research is to develop a mathematical method for recognizing a group of stress-corrosion cracks in the metal of a steel gas pipeline by the signal of a surface eddy-current sensor. Objects: groups of stress-corrosion cracks with different geometric parameters located in the metal at different intervals. Changing the geometric parameters of a group of stress-corrosion cracks and the intervals between them leads to the transformation of the signal shape of surface eddy-current sensors during non-destructive testing of a steel gas pipeline with eddy-current flaw detectors. Methods: mathematical modeling of signals of surface eddy-current sensors over a group of stress-corrosion cracks in metal; experimental studies to determine empirical dependencies for informative parameters of the signal of a surface eddy-current sensor (amplitude, signal half-width) on the interval between continuity defects on artificial models of stress-corrosion cracks; empirical method for recognizing a group of stress-corrosion cracks from measured signals of surface eddy-current sensors. Results. The paper introduces the basics of an empirical method for recognizing a group of stress-corrosion cracks in metal by the signal of an absolute surface eddy-current sensor. Based on a mathematical model of the signal of a surface eddy-current sensor the authors established the parameters (amplitude and half-widths of the signal over a group of cracks) containing consolidated information on the number of continuity defects in a group of stress-corrosion cracks and the interval between them. The empirical dependences of the signals of a surface eddy-current sensor were experimentally obtained on artificial models of stress-corrosion cracks in steel samples. The applicability of the empirical method of recognizing a group of stress-corrosion cracks in metal when using the coefficients of the wavelet transformation of the signal of surface eddy-current sensor of different types (absolute and differential) as signs of the classification of continuity defects is shown.
Keywords
