Weld joint ultrasonic phased array detection data intelligent analysis method based on deep learning

An ultrasonic phased array and data detection technology, which is applied to the analysis of solids using sound waves/ultrasonic waves/infrasonic waves, the use of sound waves/ultrasonic waves/infrasonic waves for material analysis, and material analysis. Control array detection and other issues

Active Publication Date: 2020-04-24
WUHAN WUCHUAN MEASUREMENT & TEST
View PDF6 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies in the existing technology, the present invention proposes a deep learning-based intelligent analysis method for welding seam ultrasonic phased...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Weld joint ultrasonic phased array detection data intelligent analysis method based on deep learning
  • Weld joint ultrasonic phased array detection data intelligent analysis method based on deep learning
  • Weld joint ultrasonic phased array detection data intelligent analysis method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The embodiments of the present invention are further described in detail below.

[0036] This embodiment provides a method for intelligent analysis of welding seam ultrasonic phased array detection data based on deep learning. The method includes the following steps:

[0037] S1: Perform data sampling. Select steel plates with a thickness of 12, 14, 18, 20, 24, and 30 mm, and the ratio of X-shaped welds to V-shaped welds in the steel plate welds is 1:1. The phased array is then used to inspect the welds of the steel plates, and the phased array weld slices are sampled every 1 mm to obtain a series of S-scan pictures. In order to make the number of samples sufficient and to improve the generalization ability of the deep network, scan the other side of the same weld to obtain data, such as figure 1 shown.

[0038] S2: Perform preprocessing. Divide the S-scan pictures in step S1 into a training set and a verification set according to a ratio of 1:1, and keep the propo...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a weld joint ultrasonic phased array detection data intelligent analysis method based on deep learning, and the method comprises the steps: S1, carrying out data sampling: carrying out detection on weld joints of steel plates with different thicknesses through a phased array, and obtaining a series of S-scanning pictures; S2, carrying out preprocessing: dividing the S-scanning pictures into a training set and a verification set according to the proportion of 1: 1, and resetting the sizes of the pictures to be 800*800; S3: carrying out data annotation: carrying out defect annotation on the S scanning pictures in the training set and the verification set by using annotation software labelImg; S4, training a Faster RCNN: circularly and alternately training the Faster RCNN and the RPN etwork by using the data set marked in the S3 and a pre-trained VggNet16 network weight model; and S5, carrying out result testing: inputting a to-be-detected picture into the trainednetwork, and outputting a detection result. The method not only improves the recognition rate of the S-scanning two-dimensional plane defects in the ultrasonic phased array detection of the weld joints, but also has the advantages of high accuracy and low omission ratio.

Description

technical field [0001] The invention relates to the technical field of ultrasonic phased array detection, in particular to an intelligent analysis method for weld seam ultrasonic phased array detection data based on deep learning. Background technique [0002] At present, non-destructive testing after welding is a necessary step in industrial production and the main means of testing welding quality. Ultrasonic phased array technology can provide more detection information for inspectors, and the unique working methods of ultrasonic phased array technology, such as linear scanning, sector scanning, and dynamic focusing, can be applied to more complex workpiece inspections. It has a higher defect detection rate and is generally used for the detection of welding products with higher quality requirements. Ultrasonic phased array detection technology will form five basic views in the final detection imaging: A-scan view, B-scan view, C-scan view, D-scan view and S-scan view, whe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G01N29/06G01N29/44
CPCG01N29/0654G01N29/4481G01N2291/267Y02P90/30
Inventor 李威王旭之朱甜甜王礼宾宋波
Owner WUHAN WUCHUAN MEASUREMENT & TEST
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap