Automation of the Choice of the Welding Processes by a Neuronal Approach

In this study, we propose a new model for the automatic choice of the welding process based on the use of the multi-layers artificial neural networks (MLANN). The automatic choice of welding process is done in three steps; firstly, we started with the formulation of a training base which represents the experiments of the experts of the field as well as the necessary knowledge of know-how, and which is among the general criteria for the choice of welding process. The second step was devoted to the creation of the first multi-layers neural networks NN1 for the choice of the assembly mode or four modes are presented, whereas the finally step is relate to the choice of welding process.