Affiliation:
1. Southern Methodist University Research Center for Advanced Manufacturing Richardson, Texas, USA
Abstract
This paper presents a technique to monitor the weld penetration in real time by the front-side image sensing of the keyhole welding pool in variable polarity plasma arc welding (VPPAW) of a 6061 aluminium alloy. A novel optical system is developed to acquire the keyhole welding pool images from two sides of the plasma arc welding torch. The geometric sizes of the visible keyhole can be accurately obtained by using the developed image-processing algorithm. The variation in sizes of the visible keyhole has a close relationship with the real diameter of the keyhole. An artificial neural network (ANN) model is established to describe the relationship between the keyhole diameter and the geometrical sizes of the keyhole welding pool obtained from the image.
Subject
Industrial and Manufacturing Engineering,Mechanical Engineering
Cited by
11 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献