A method for detecting welding quality of component arc welding

Abstract

The invention provides a component arc welding welding quality detection method, which firstly collects a certain number of pictures of multiple different types of welding heads of the same component arc welding, and establishes a sample library. Part of the sample library is used as a training set, and the other part is used as a test set; the template matching algorithm based on the spatial correlation method extracts the target welding joint area pictures in each sample in the training set and the test set; The welding head quality types of all pictures are manually marked, and the convolutional neural network is built based on the tensorflow framework, and all the target welding head area pictures of each category in the training set are input into the convolutional neural network to train the machine learning classifier; finally, the S test set The pictures of the target welding joint area in each sample image are classified by the trained machine learning classifier, and the classification result is the corresponding welding joint quality type. The invention can quickly discover the quality problems of arc welding of components in the picture.

Description

technical field [0001] The invention relates to the technical field of welding quality detection, in particular to a method for detecting welding quality of component arc welding. Background technique [0002] The welding quality inspection of components refers to the inspection of welding results, the purpose is to ensure the integrity, reliability, safety and usability of the welding structure. [0003] Traditional manual inspection needs to be inspected according to welding structure instructions, technical standards, process documents, construction drawings, etc. The inspection of appearance quality mainly includes dimension inspection, geometric shape inspection, and appearance flaw inspection. Radiographic flaw detection and ultrasonic testing can be used Flaw detection, coloring inspection and other methods for auxiliary detection. [0004] However, there are many deficiencies in the traditional manual detection and classification. Only relying on human eyes and some...

AI Technical Summary

Problems solved by technology

[0004] However, there are many deficiencies in the traditional manual detection and classification. Only relying on human eyes and some inspection rules for detection and classification will inevitably lead to many errors and consume a lot of time and labor on repetitive work.

Auxiliary detection methods such as radiographic flaw detection may have higher costs, so efficient and intelligent component welding quality detection and classification methods have become an important demand in current actual production

Images