文献类型：期刊文献

英文题名：Three-dimensional vision applications in GTAW process modeling and control

作者：Wang, XueWu[1]

机构：[1]E China Univ Sci & Technol, Key Lab Adv Control & Optimizat Chem Proc, Minist Educ, Shanghai 200237, Peoples R China

年份：2015

卷号：80

期号：9-12

起止页码：1601

外文期刊名：INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY

收录：;EI(收录号：20151700794475);WOS:【SCI-EXPANDED(收录号:WOS:000361628900011)】；

基金：This work is funded by the Shanghai Natural Science Foundation under Grant 14ZR1409900. The support from Professor YuMing Zhang and assistance from PhD students Mr. WeiJie Zhang and Mr. YuKang Liu in experiments at the University of Kentucky Welding Research Laboratory are greatly appreciated.

语种：英文

外文关键词：GTAW; Sensing; Modeling; Control; Artificial intelligence

摘要：In gas tungsten arc welding (GTAW), the weld pool is the major source of information that can be used to assure the production of the desired weld penetration-the most critical factor determining the weld integrity. To meet this challenge, various sensing technologies, modeling methods, and control strategies have been studied, and artificial intelligence technologies were applied to improve system intelligence. The GTAW process analysis is given first. Then, a short introduction on weld pool sensing technologies is presented, where three-dimensional (3D) vision sensing is a very active orientation. Furthermore, weld pool description model and characteristic parameter model are also discussed, where intelligent algorithms were used generally. Besides, dynamic modeling and penetration control strategies, especially intelligent control strategies are presented. At last, the discussion about development of the GTAW penetration control product is analyzed briefly.