Determining the Tensile Strength of Parts in Gas Tungsten Arc Welding with Controlled Feeds Using Artificial Neural Networks

Gas tungsten arc welding (GTAW), is one of the important methods used in industry and it is an arc welding process that uses a no consumable tungsten electrode to produce the weld. GTAW is most commonly used to weld thin sections of stainless steel and non-ferrous metals such as aluminum, magnesium, and copper alloys. In this research the authors have designed and manufactured an automatic welding arm in which its feed is controlled by a microcontroller. In different conditions like feed, ampere and pre-heat temperature practical tests have done and tensile strength of the welded parts has measured. Then by using this data an artificial neural network designed and tested. The inputs of the network are adjustable parameters of the welding process and the output is the tensile strength. The results show that by using neural networks we can determine the tensile strength precisely before the process.