A Study on Global and Cluster-wise Regression Model in the Automatic GMA welding
-
Published:2019-10-01
Issue:1
Volume:635
Page:012029
-
ISSN:1757-8981
-
Container-title:IOP Conference Series: Materials Science and Engineering
-
language:
-
Short-container-title:IOP Conf. Ser.: Mater. Sci. Eng.
Author:
Yun Tae-Jong,Shim Ji-Yeon,Kim Hong-Gun,Liang Zong-Liang,Won-Bin Oh,Lee Bo-Ram,Park Min-Ho,Kim Ill-Soo
Abstract
Abstract
The automatic welding system is presently made use of high volume production industries even if the cost of the related equipment is justified by the large number of pieces to be made. Also, the detailed movement devices with the predetermined sequences of welding parameter and the use of timers to form the weld joints were required. A new mathematical model that predict the optimal welding parameters on a given bead geometry and accomplish the desired mechanical properties of the weldment to make the automatic GMA (Gas Metal Arc) welding process should be needed. The developed model should be employed a wide range of material thicknesses and be applicable for all welding positions as well. In addition, the algorithm must be available in the form of mathematical equations which can be programmed easily to the robot and give a high degree of confidence in predicting the bead dimensions. In this study, two regression models with global regression and cluster-wise regression are proposed to be applicable for prediction of optimal welding parameters on the bead reinforcement area. For development of the proposed regression models, an attempt has been done for applying to a several methods. A series experiments to research the effects of welding parameters on bead reinforcement area as a function of key output parameters for the lab-joint weld in the automatic GMA welding process was performed. Not only the fitting of these models was checked and compared by using a variance test (ANOVA), but also the prediction of bead reinforcement area using the developed regression models were carried out the basis of the additional experiments.
Reference10 articles.
1. Welding automation and computer control;Marburger,1990
2. The geometry of gas tungsten arc gas metal arc and submerged arc weld beads;Heipe,1990
3. Prediction of laser butt joint welding parameters using back-propagation and learning vector quantization networks;Jeng;Journal of Materials Process Technology,2000
4. Study on prediction of optimized penetration using the neural network and empirical models;Kim;Journal of Korean Society of Machine Tool Engineers,1999
5. Neural networks for online prediction of quality in gas metal arc welding;Li;Science and Technology of Welding and Joining,2000