Defect classification method and system based on ultrasonic phased array imaging

Abstract

The invention provides a defect classification method and system based on ultrasonic phased array imaging, and belongs to the field of nondestructive detection.The method comprises the steps that ultrasonic phased array imaging is adopted for a to-be-detected sample piece to obtain ultrasonic full-matrix data; performing full-focusing processing on the ultrasonic full-matrix data, performing color coding according to the signal amplitude, and obtaining an image of the sample piece to be detected; preprocessing an image of a to-be-detected sample, inputting the preprocessed image into the classification prediction model, and obtaining defect classification of the to-be-detected sample; a method for training the classification prediction model comprises the following steps of: acquiring simulation data by simulating ultrasonic phased array imaging by utilizing a defect ultrasonic scattering data finite element simulation method; after full-focusing processing is carried out on the simulation data, a simulation image is obtained; preprocessing the simulation image and then performing data enhancement; carrying out image feature extraction on the simulation image by adopting a convolutional neural network; and inputting the image features into a full connection layer, and training a classification prediction model by taking defect classification as output. According to the invention, the accuracy of defect classification is improved.

Description

technical field [0001] The invention belongs to the field of nondestructive testing, and more specifically relates to a defect classification method and system based on ultrasonic phased array imaging. Background technique [0002] The development and application of ultrasonic nondestructive testing technology is based on the interaction between ultrasonic waves and the object to be measured; when ultrasonic waves with good guidance encounter defects during propagation, their propagation direction or characteristics will change. The study of reflection, refraction and scattering can realize the detection and characterization of workpiece defects. Compared with other non-destructive testing methods, ultrasonic non-destructive testing has unique advantages; it is widely applicable to non-destructive evaluation of metals, non-metals and composite materials; it has strong penetration ability, large detection depth, and good positioning ability for internal defects of workpieces;...

AI Technical Summary

Problems solved by technology

[0003] In the existing ultrasonic defect detection technology, there is a lack of detection of air hole defects, that is, the lack of parameterization data of multiple small and close holes

In addition, conventional B-ultrasound imaging has low resolution and poor imaging accuracy, and has no ability to resolve multiple small defects (size less than 0.8 wavelength) in a short distance

Among common classification methods, dimensionality reduction algorithms such as linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) can simplify high-dimensional data to deal with large-scale data sets, but it is difficult to understand the meaning of the results; Naive Bayesian method Classification is based on probability theory, so this method is limited by Bayesian theorem and conditional independence assumption, but the conditional independence assumption is often not established in practical applications, thus affecting the classification effect of Bayesian method

Images