Font Size: a A A

Research On Defect Type Recognition Of Metal Pipeline Butt Weld By Ultrasound Detection

Posted on:2021-05-31Degree:MasterType:Thesis
Country:ChinaCandidate:H X RenFull Text:PDF
GTID:2381330614456276Subject:Safety engineering
Abstract/Summary:PDF Full Text Request
Ultra supercritical boiler has remarkable energy-saving effect and high thermal efficiency,which is in line with the concept of green,low-carbon and efficient development of our country.However,the tube burst accident caused by the weld defect of the water-cooled wall of the spiral tube ring of the boiler can not be ignored,and each shutdown will bring huge human,material and financial losses.Therefore,the defects in the weld of the water-cooled wall fins are effectively identified,and then corresponding measures are taken It is of great significance to save resources,ensure normal operation of boiler and reduce accidents.Ultrasonic has the function of transmitting energy and information,so ultrasonic detection technology is widely used.At present,ultrasonic detection technology can successfully locate the location of defects and calculate the size of defects,but there are still some difficulties for defect identification.In this paper,the cracks,holes,incomplete penetration and incomplete fusion in the water wall fin weld of ultra supercritical boiler are studied.The defects are collected by ultrasonic detection.Firstly,the defect signal is truncated and denoised by wavelet,then the defect signal is analyzed by Fourier transform,short-time Fourier transform and wavelet transform,and different defect signals are extracted The characteristic value of the.Through comparative analysis,it is found that the power spectrum density and energy spectrum of different defect signals are different,so these two eigenvalues can be selected as the basis of pattern recognition,and then the quantitative recognition of defects can be achieved.
Keywords/Search Tags:Ultra Supercritical Boiler, Water wall weld of spiral pipe ring, Ultrasonic testing, Time frequency analysis, pattern recognition
PDF Full Text Request
Related items