Large structural part variable position welding quality online monitoring method and system

Abstract

The invention provides an online monitoring method and system for variable-position welding quality of a large structural part, and belongs to the technical field of intelligent welding manufacturing. In order to solve the problem that an efficient welding quality detection means is lacked in automatic welding of existing large structural parts, real-time collection and stable wireless transmission of molten pool images in the variable-position welding process of the large structural parts are achieved through passive vision sensing and WiFi (Wireless Fidelity) wireless broadband communication technologies; on-line monitoring of typical welding seam forming defects is achieved based on a machine learning classification model by extracting molten pool multi-target domain features in a continuous molten pool image sequence. According to the method, a basis can be provided for in-situ nondestructive testing and in-situ rapid repair welding after welding, the defect detection efficiency is greatly improved, meanwhile, the rejection rate is reduced, the downtime is shortened, then the time and cost of the whole welding manufacturing period are reduced, and the method is particularly suitable for large-scale welding manufacturing of large structural parts such as aerospace propeller storage box end sockets.

Description

technical field [0001] The invention relates to the technical field of intelligent welding manufacturing. In particular, it relates to an online monitoring method and system for variable position welding quality of large structural parts, which can be widely used in the welding manufacture of large structural parts such as storage tanks, storage tanks, and pressure vessels. Background technique [0002] With the large-scale development of mechanical equipment in aerospace, large ships, nuclear power, clean energy and other fields, higher requirements are put forward for intelligent welding and welding precision. In the welding and manufacturing process of large and complex components, due to the large weight and volume of the components, it is difficult to move and turn over, and it is difficult to apply the traditional welding production method of "positioner-assisted to ensure the flat welding position". The welding method with continuously changing position (that is, var...

AI Technical Summary

Problems solved by technology

The above technical solutions only involve the extraction of molten pool characteristic parameter information and electrical parameter information and welding quality monitoring in traditional welding methods such as flat welding and vertical welding. In addition, the existing technical solutions only involve welding process information such as molten pool images and electrical parameters. There is no real-time acquisition and wireless transmission of molten pool images and intelligent online monitoring of welding quality for large-scale structural parts variable position welding.

Images