Return to search

VISION BASED REAL-TIME MONITORING AND CONTROL OF METAL TRANSFER IN LASER ENHANCED GAS METAL

Laser enhanced gas metal arc welding (GMAW) is a novel welding process where a laser is applied to provide an auxiliary detaching force to help detach the droplet such that welds may be made in gas tungsten arc welding high quality at GMAW high speeds. The current needed to generate the electromagnetic (detaching) force is thus reduced. The reduction in the current helps reduce the impact on the weld pool and over-heat fumes/smokes. However, in the previous studies, a continuous laser is applied. Since the auxiliary is only needed each time the droplet needs to be detached and the detachment time is relatively short in the transfer cycle, the laser energy is greatly wasted in the rest of the transfer cycle. In addition, the unnecessary application of the laser on the droplet causes additional over-heat fumes. Hence, this study proposes to use a pulsed laser such that the peak pulse is applied only when the droplet is ready to detach. To this end, the state of the droplet development needs to be closely monitored in real-time. Since the metal transfer is an ultra-high speed process and the most reliable method to monitor should be based on visual feedback, a high imaging system has been proposed to monitor the real-time development of the droplet. A high-speed image processing system has been developed to real-time extract the developing droplet. A closed-loop control system has been established to use the real-time imaging processing result on the monitoring of the developing droplet to determine if the laser peak pulse needs to be applied. Experiments verified the effectiveness of the proposed methods and established system. A controlled novel process – pulsed laser-enhanced GMAW - is thus established for possible applications in producing high-quality welds at GMAW speeds.

Identiferoai:union.ndltd.org:uky.edu/oai:uknowledge.uky.edu:ece_etds-1017
Date01 January 2013
CreatorsShao, Yan
PublisherUKnowledge
Source SetsUniversity of Kentucky
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceTheses and Dissertations--Electrical and Computer Engineering

Page generated in 0.0021 seconds