The Neural Network Algorithm for Decision Support During Control of the Welding Process by Robotic Complexes

The article proposes an algorithm for solving the problem of controlling the welding process in robotic technological complexes according to the quality criterion of the products. The statement of the problem is given and the model of system dynamics by J. Forrester is considered. Welding quality indicators in robotic complexes were used as system levels. Functional dependencies are taken by using the mathematical apparatus of artificial neural networks, in particular, radial basis functions that allow for rigid interpolation. The introduction of the models and algorithms discussed in the article will improve the quality of products using robotic technological complexes.