Additive manufacturing weld penetration online detection and control method and system

Abstract

The invention discloses an additive manufacturing weld penetration online detection and control method and system. The method comprises the steps that a to-be-welded part is welded according to process parameters; infrared images of a molten pool are collected, the molten pool images are subjected to image processing, and the weld width of the molten pool part and the maximum temperature difference in the weld penetration direction are obtained; a theoretical weld penetration calculation module, calculation parameters are input into the theoretical weld penetration calculation module, and theoretical weld penetration is obtained; the theoretical weld penetration, the weld width and the maximum temperature difference in the weld penetration direction are input into a trained artificial neural network, and actual weld penetration is output through the artificial neural network; and a difference value between the actual weld penetration and preset required weld penetration are calculated,when the difference value exceeds a preset range, the process parameters are adjusted according to the difference value, and then a next to-be-welded part is welded. Real-time online detection of theweld penetration is achieved, the weld penetration is controlled online in real time, intelligent additive manufacturing is achieved, and the automation degree is high.

Description

technical field [0001] The invention relates to the field of welding penetration detection and control, in particular to an online detection and control method and system for additive manufacturing penetration. Background technique [0002] The concept of additive manufacturing began in the late 1980s, and my country began to study it in the early 1990s. After just over 20 years, this technology has achieved rapid development, and has broad application prospects in many fields such as aerospace, micro-nano manufacturing, and biomedical engineering. The penetration depth in arc additive manufacturing is an important parameter related to welding quality. Research on online real-time control of penetration is of great significance for the development of automation and intelligence in additive manufacturing. [0003] An ultrasonic phased array method for measuring U-rib weld penetration and an ultrasonic penetration detection system disclose methods for detecting weld penetrat...

AI Technical Summary

Problems solved by technology

[0003] An ultrasonic phased array method for measuring U-rib weld penetration and an ultrasonic penetration detection system disclose methods for detecting weld penetration using ultrasonic waves, but ultrasonic detection requires contact between the device and the workpiece, which affects the dimensional accuracy of the product; At the same time, as the position of the welding seam moves, the device also needs to move, which makes it difficult to achieve process synchronization and intelligent manufacturing

A weld penetration detection method discloses a method for detecting penetration using changes in magnetic induction signals, which is difficult to use for high-temperature weld detection

[0004] At present, there is no method for online real-time detection and online real-time control of weld penetration

Images