EFFECT OF WELDING PARAMETERS ON BEAD GEOMETRY AND METALLURGY IN SAW PROCESS USING CCD

Abstract

Submerged Arc Welding is the most extensively used permanent joining process for joining of thick sections. The most important features of this process are high deposition rate, ability to weld thick sections with ease and longer weld runs. Effect of process parameters: Arc voltage, Welding current, Travel speed and Nozzle-to-plate distance, on the weld process is determined by co relating the process parameters with bead geometry features such as, bead reinforcement height, bead width and bead penetration. The design matrix was prepared on the basis of 4 factors, 5 levels, rotatable Central Composite Design. Response Surface Methodology was used to develop the mathematical models co relating the process parameters with the bead geometry features. The models once developed were checked for adequacy using ANOVA technique. From the adequate models the significant terms were selected using p test. The finally proposed models contains only the significant terms. Main and interaction effects of the process variables on weld bead geometry are presented in graphical form. The developed models can be used for prediction of important weld bead dimensions and control of the weld bead quality by selecting appropriate process parameter values. Use of artificial neural networks for modelling of the Submerged Arc Welding process was done. Artificial Neural Network architecture, using back propagation algorithm was developed which provided satisfactory outputs. Comparison of the performance of the RSM models and the ANN model was also done and it was concluded that the when number of factors are less, RSM yields more satisfactory results. v

Metallurgical investigations determining the variation of micro hardness across the weld metal zone, heat affected zone and the base metal were carried out. Effect of parameters on the variation of the Knoop’s micro hardness is determined. Also the microstructure of the resultant welded metal was co-related with the process variables. This thesis is divided in 8 chapters. The first chapter discusses the objective and motivation of the project, followed by the statement of problem and lastly the plan of investigation, which was undertaken to achieve the objectives. The next chapter gives an insight regarding the research work which has been carried out in related fields such as ANN modeling, RSM application to welding problems, metallurgical investigations etc. Chapter 3 gives a brief introduction to the concepts of SAW, ANN, RSM and weld metallurgy, followed by the experimental procedures undertaken to carry out the project work. Chapter 4 discusses the development of mathematical models. It is followed by Chapter 5 where in discussions of the effects of process variables on bead geometry parameters are done. Chapter 6 is dedicated to use of ANN for modeling of the SAW process. A comparison of the 2 modeling approaches is also done. Chapter 7 discusses the metallurgical investigations in micro hardness and micro structure. Chapter 8 states the results and conclusions drawn from this study.