A welding quality evaluation method and system based on ultrasonic image feature fusion

Abstract

The invention provides a welding quality evaluation method and system based on ultrasonic image feature fusion, and relates to the technical field of welding quality detection. Multiple defect feature quantities to achieve comprehensive characterization of welding quality, comprehensive evaluation information and high accuracy; the method includes: S1. Perform ultrasonic testing on backfilled friction stir spot welding joints, and collect B-scan images and C-scan images of the welding area. Scan the image; S2, analyze and process the obtained B-scan image and C-scan image, extract the effective information in the image and the defect features inside the solder joint, and form a three-dimensional information representation of the welding quality; S3, use the information entropy to analyze the three-dimensional information. The feature quantities in the representation are given different weight coefficients, and then merged into a comprehensive index as the evaluation result. The technical solution provided by the present invention is suitable for the process of welding quality evaluation.

Description

technical field [0001] The invention relates to the technical field of welding quality detection, in particular to a welding quality evaluation method and system based on ultrasonic image feature fusion. Background technique [0002] Backfill friction stir spot welding is a new type of welding technology. Because there is no keyhole after welding, the joint has high strength and good quality, it is widely used in the welding of commonly used light metals such as aluminum alloy and magnesium alloy in aerospace, automobile, shipbuilding and other fields. However, in the welding process, improper selection of welding parameters will lead to insufficient backfilling, hook defects, holes and other defects inside the joint. These defects are often easy to expand into cracks, thereby reducing the mechanical properties of the product. Therefore, the detection of welding defects and the evaluation of welding quality are very important. [0003] Traditional welding quality detection...

AI Technical Summary

Problems solved by technology

This patent is mainly for the analysis of A-scan waveforms. The larger the scanning range, the more A-scan waveforms will be obtained, which will lead to problems such as large amount of analyzed data and low analysis efficiency.

More importantly, the A-scan waveform is a one-dimensional signal, and it is difficult to fully characterize the distribution characteristics of welding defects in three-dimensional space, so the ability of this technology to evaluate welding quality is limited

Images