| Defect detection has always been a hot topic in real-life production.Whether it is in the pipeline inspection of natural gas or petroleum,or the parts of a ship or automobile,defects cannot be detected.At present,the most widely used defect detection technology in industry is non-destructive testing,and the most widely used detection method in non-destructive testing technology is ultrasonic testing.The utility model can detect the workpiece without destroying the internal structure of the detected object,and detects defects such as impurities,bubbles,holes,and cracks in the detected object,so that the worker can better analyze and improve the performance of the product.,increase product rate and avoid economic losses and safety accidents caused by damaged parts.At present,the defect detection is performed on the inside of the workpiece by using the ultrasonic detection method,and defect information in the workpiece can be effectively detected.Due to the diffusion of ultrasonic beam and the increase of angle,the directivity of acoustic beam becomes worse as the acoustic propagation distance increases.This paper focuses on this issue,taking into account the influence of different spatial resolution on the reconstruction of the ultrasonic echo data from the ultrasonic echo data.Starting from the characteristics,two solutions are proposed in the reconstruction of defects: point cloud method and surface cloud method.The former is the basic solution to the defect reconstruction in the case of high spatial resolution of ultrasound,and the latter is the reconstruction measures taken when the spatial resolution is low.The face cloud method considers that the reflection of each reflection point from different incident waves is anisotropic and reconstructs the defect type based on the reflection characteristics.The two methods acquire the ultrasonic echo data by combining the full matrix capture(FMC)method with the simulation of four different defects.The Hilbert transform is used to process and analyze the echo signals,and the total focus method(TFM)principle is used.The post-processing of the echo data is performed.Finally,point cloud distribution with different defects is realized in MATLAB.The face cloud method also calculates the echo reflection angle according to the law of ultrasonic reflection.Finally,the interpolation method is used in MATLAB according to the angle and combined with finite element analysis.Approximate functions are used to express the position equation of the surface to achieve the reconstruction of different defect surfaces.The dissertation focuses on the reconstruction of defects in ultrasonic testing.Based on the analysis of the reconstruction effect of point cloud method,the method of surface cloud reconstruction is proposed.By finding the main reflection surface of the defect,a relatively high reconstruction resolution is obtained under the premise of coarser mesh division.This will help to improve inspection quality and inspection efficiency. |
