Model development of resistance spot welding on low carbon steel and advanced high strength steel using multiresponse DOE and FEM / Norasiah Muhammad

Muhammad, Norasiah (2015) Model development of resistance spot welding on low carbon steel and advanced high strength steel using multiresponse DOE and FEM / Norasiah Muhammad. PhD thesis, Universiti Teknologi MARA.

[img]
Preview
Text
TP_NORASIAH MUHAMMAD EM 15_5.pdf

Download (1MB) | Preview

Abstract

This project investigated the development of the weld zone in resistance spot welding (RSW) on similar and dissimilar material using various statistical methods and numerical simulation approach. Demands placed on dissimilar material joints have increased from various viewpoint such as lightweight vehicle structure, energy saving, high performance and cost saving. Due to steadily increasing price of material, a combination between DP 600 and low carbon steel can be seen as an alternative especially for the section which is not fully exposed to external load. Therefore, the RSW process on dissimilar material of DP 600 and low carbon steel is necessarily to be observed for mass production. For comparison purpose, this research was conducted for plate thickness of 1.2 mm of low carbon steel and 1.0 mm of DP600. In order to reduce the trial and error experiments, statistical methods were used throughout the investigation. An experimental design was started using a general 24 factorial design to determine which of the various parameters were important in response surface study. The significance of the welding parameters was obtained using Analysis of Variance (ANOVA) to evaluate the relations existing between the important parameters and the response (radius of weld nugget and radius of HAZ). Optimization of the welding parameters (weld current, weld time and electrode force) to normalize weld nugget and to minimize HAZ size was then conducted using Central Composite Design (CCO) in Response Surface Methodology (RSM) and the optimum parameters were determined. Experiments were conducted according to the selected experimental design, followed by data analysis which included regression analysis and model adequacy checking. A quadratic model for radius of weld nugget and radius of HAZ as a function of the significant parameters were developed for all the combined metal sheets. A second order models were found fitted and can be effectively used to predict the size of weld zone within the factors domain study. Further, the optimization of multi-response (nominal-the-best for radius of weld nugget and smaller-the-better for radius of HAZ) were obtained simultaneously using desirability approach. The confirmation tests validated the use of multi-response optimization for enhancing the welding performance and optimizing the welding parameters in RSW process. The experimental obtained under the optimum operating conditions and the predicted one was found to agree satisfactorily with each other. The optimum parameters obtained from multi-response optimization was applied in the simulation process. A two dimensional axis-symmetric finite element model using customized electrode meshing was chosen to develop the thermal-mechanical-electrical characteristic of the RSW based on the actual electrode dimension. The development of the weld zone was investigated using finite element method (FEM) of SYSWELD and the results were compared with experimental measurement. It was found that the result of FEM presents a theoretically accurate correlation with that of experiments in terms of radius weld nugget as well as HAZ developed.

Item Type: Thesis (PhD)
Uncontrolled Keywords: Resistance spot welding; Low Carbon Steel; Strength Steel; Multiresponse DOE and FEM
Subjects: T Technology > TS Manufactures > Metal manufactures. Metalworking > Welding
Divisions: Faculty of Mechanical Engineering
Depositing User: Staf Pendigitalan 5
Date Deposited: 20 Mar 2017 02:45
Last Modified: 20 Mar 2017 02:45
URI: http://ir.uitm.edu.my/id/eprint/16263

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year