Study On Brazing Detection Method For Stator Blade Of High Pressure Compressor

The high pressure compressor is an important part of the aero-engine.As an important part of the compressor,the stator blade is now manufactured by brazing.The quality of welding directly affects the overall performance of the engine.Since the component is welded,it is difficult to use the conventional NDT,so it's necessary to develop a non-destructive testing method that can be online.This paper develops a potentiometric detection method for the brazing seam of the third to sixth-grade brazed seam of the high-pressure compressor stator blades.Firstly,Using ANSYS simulation software,the appropriate detection current and the appropriate size of the detection electrode had been studied and the defect size suitable for detection had been studied.Brazed analog blades and specimens were designed and fabricated.Using ultrasonic C-scan and ray DR to calibrate the welding quality.The probes and fixtures were designed and manufactured to meet the requirements of different test specimens.Brazed analog and specimens were detected and analysed.The results of the research show that the analysis results of sample data are in good agreement with the results of ANSYS simulations.The results of actual ultrasonic C scans also show good similarities.The DR detection value of the blade and the value of the potentiometric detection method are linear.The fitting function is y=168.67201+ 192.44305 x.It has been verified that increasing the detection voltage(current)has little influence on the detection sensitivity,and the use of a small electrode is more conducive to the detection of defects,and It is difficult to detect defects with a large defect size.Based on this,a decibel line analysis method using the kernel density function approximate function(normal function)peak decline is proposed for the on-site detection requirements.Using MATLAB to draw the box plot of the measurement data.Remove the corresponding outliers based on the box line drawing and the actual condition of the vane,and use the Jarque Bera test,Calori test and other methods to predict the distribution of sample data.Based on the distribution of sample data,the nuclear density function of each class of blade is estimated.the distribution regularity and the nuclear density estimation function of each stator blade are obtained,and then the nuclear density function and clustering are used.The normal range of the detection resistance value is drawn by the density function and the clustering method.the detection range is corrected in conjunction with the actual condition of the blade.The results show that the quality of brazing can be judged according to the peak value of the nuclear density function and the cluster analysis.The best indicators of welding are the peaks of 6db and 8db of nuclear density detected by single and double blade three-grade blades;The peak density of double-blade detection decreased by 8db;the peak density of single and double-blade detection of five-grade blade decreased by 9db and 6db;the peak value of nuclear density detected by single-blade and double-blade detection decreased by 8db and 9db.The area outside the detection line is Areas that require focus,this result has been successfully validated on the blades.The research results of this project provide an effective detection method for the detection of brazing of a certain type of high-pressure compressor rectification vanes,which provides a basis for the development of potentiometric inspection standards,processes,and acceptance criteria,and ensures the rectification of a new compressor.Development quality.