Nowadays, one of the most popular welding processes is Tungsten Inert Gas -TIG- due to its high quality and slag free joints. Joining thin sheet and pipe metals are one of the most important applications of this process. The effects of TIG welding process parameters on ultimate tensile strength -UTS- of thin pipes have been investigated. The material used for specimens was AISI 304 stainless steel pipes which is excessively used in oil and gas and power plant pipelines. Taguchi design of experiments -DOE- approach has been performed to design the experiment matrix and gathering required data for modeling and optimization. Different regression models have been used to establish the relationship between input and output parameters. Current -I-, frequency -F-, welding speed -S-, gap -G- and debi -D- are the most important input parameters in TIG welding process considered in this paper. Ultimate tensile stress -UTS- has been selected as the quality characteristic to assess the process. The most fitted model was selected using analysis of variance -ANOVA-. The UTS model was then maximized using simulated annealing -SA- algorithm. The performance of the proposed optimization method has been proved through experimental test.

Keywords