Research On Key Technology Of Non-contact Automatic Detection Of Butt Weld

As an important processing and forming method,welding is widely used in various industries,such as petrochemical,nuclear industry,ships and bridges.Welds are important joints between different structures,but for a variety of reasons,welds and nearby areas are often the most prone to defects.In the weld,butt welds account for the majority.At present,the methods for weld inspection are mainly magnetic powder,penetration and conventional ultrasonic testing.These methods are time-consuming and labor intensive,which is not conducive to rapid and automatic detection.Exploring a non-contact detection method,the contact surface is not added with a coupling agent,and the scanning speed is fast,so as to realize the automatic detection of the weld seam,which is very meaningful for the automatic detection technology of the butt weld.In this paper,the key technologies,transducer design and signal noise reduction methods are studied for the automatic detection of butt welds.Firstly,the transducer parameters are optimized by finite element software,and the external structure of the transducer is designed in combination with the actual needs of the automatic detector.After that,the experimental platform was built to test the transducer and the signal noise reduction algorithm was studied.Finally,an automatic detector for butt welds is designed to explore the method for automatic detection of butt welds.The main contents are as follows:(1)Firstly,based on the Lorentz force mechanism and magnetostrictive mechanism of ferromagnetic materials,the excitation principle of the transducer is analyzed.According to the mechanism analysis,the permanent magnets and the tortuous coils that provide the vertical magnetic field are matched with the unique excitation frequency and the spacing of the coil turns,so that oblique ultrasonic waves of a certain angle can be excited in the test piece.Finally,the common defects in the weld are analyzed,and the characteristics of the weld inspection are combined to determine the oblique incident transducer.(2)According to the structure of obliquely incident Shear Vertical(SV)wave transducer,a two-dimensional finite element model is established to analyze the fixed magnetic field and vibration displacement in the specimen.By analyzing the fixed magnetic field generated by the permanent magnet in the test piece,it is found that the closer the permanent magnet is to the magnetic field,the stronger the magnetic field is;and even if the magnetic field direction of the permanent magnet is perpendicular,a horizontal magnetic field is generated in the test piece,and the horizontal magnetic field line is distributed.On both sides of the permanent magnet and in opposite directions;the magnetization effect of the permanent magnet on the test piece is concentrated in a small range,the magnetic field is the strongest at the edge of the permanent magnet,and evenly distributed under the coverage,the magnetic field strength is basically the same,mainly for The magnetic field underneath.By analyzing the vibration displacement cloud diagram in the test piece,it can be seen that the vibration energy generated in the test piece is mainly concentrated at a specified angle.Extracting the vibration displacement curve at the reference point to calculate the speed of sound,which is basically consistent with the shear wave velocity,proves that the required waveform is actually excited.(3)Analyze the influence of different permanent magnet parameters on the excitation efficiency,obtain the vibration displacement curve of the reference point,and use the curve amplitude as the evaluation criterion to evaluate the excitation efficiency of the transducer.Firstly,the influence of the residual magnetic flux density of the permanent magnet is analyzed,and it is found that the amplitude increases as the residual magnetic flux density of the permanent magnet increases.Then the orthogonal test table is designed to optimize the structural parameters of the permanent magnet: width,height and lift distance.The variation trend of the amplitude with several parameters is obtained,and the optimal parameter combination is determined.(4)Analyze the influence of coil parameters in the same way as permanent magnets.Firstly,the influence of the number of turns of the coil on the amplitude of the vibration displacement at the reference point is analyzed,and the amplitude increases with the increase of the number of turns.The orthogonal test table is designed,and the coil structural parameters: width,thickness,line spacing and lift distance are optimized.The influence of several parameters on the amplitude is obtained,and the optimal combination of coil structure parameters is determined.Finally,according to the needs of automatic detection,the complete structure of the transducer is designed,and the wear-resistant layer and the roller are automatically assisted.(5)According to the results of the third chapter transducer parameter optimization finite element analysis and the structural design of the transducer,two kinds of transducers are produced for experimental test to verify the effect of transducer parameter optimization,and the final improvement effect is up to 124%.According to the calibration method of piezoelectric ultrasonic,the sound velocity of the transducer is calibrated,and the final result is only 2.25% different from the theoretical sound velocity.The weld test panels with different types of defects are processed and scanned,and the echo signal characteristics of the transducer when scanning cracks and slag inclusion defects are analyzed.(6)The noisy signal is constructed,and the noise reduction effects of several different noise reduction methods are compared.The signal-to-noise ratio and root mean square error are used as evaluation indexes to optimize the optimal combination among several methods.After determining the optimal combination,several methods are used to verify the effect of the actual echo signal denoising process.The results show that the method based on the singular value is the worst,and the wavelet threshold denoising method can remove the noise but reduce the useful component amplitude.It is proposed to use the method of combining empirical mode decomposition and wavelet threshold method to process echo signals,which can keep the useful components of echo signals from loss in the case of good noise removal.(7)Combined with the characteristics of the butt weld of the flat plate,the hardware structure of the automatic tester was designed,and the function of the remote control system was planned according to the actual test needs.Combined with the transducer designed in the previous article,the prototype of the detector is produced,the simulation test board is processed,and the test is carried out.The final result proves that the detector can complete the automatic detection and generate the detection map with the defect mark.