Good weld quality requires control of weldpool width and cooling time. The cooling time t 8/5 (from 800 °C to 500 °C) is related to the weld metal hardness. The weld metal is affected by rapid heating and rapid cooling, with various metallographic structures being formed. An important factor is appropriate selection of the cooling time to obtain a good structure and suitable strength after welding. The cooling time is strongly affected by the heat input (arc voltage U, welding current I, welding speed S, and heat input efficiency η).Numerical simulations of TIG pipe welding were initially performed by the finite difference method on the basis of heat conduction theory. The weldpool width and cooling time were then calculated at various welding currents and welding speeds. A weldpool model was constructed by a neural network using the above data as teaching signals. The welding speed and welding current were determined to obtain the required weldpool width and cooling time using the neural network model.

No abstract

This paper deals with the problem concerning the weld pool width and the cooling ratc in circumfcrcntial GTA welding of pipe. In order to obtain thc high quality of the weld, it is important to keep the weld pool width and the cooling ratc constant. The numerical simulations in the circumferential GTA welding of pipe are performed by using the finite difference method. A weld pool model is constructed by neural network. The welding speed and the wclding current are dctermined to obtain the desired pool width and the cooling rate by using the neural network model and the steepest descent method. A new approach is proposed that the welding is made clockwise and countcrclockwise simultaneously from 0' of pipe by using twin heat source.

No abstract