Modelowanie stanu warstwy wierzchniej za pomocą sieci neuronowych

The article presents result of residual stresses measured in the samples which were firstly subjected to peripheral milling. The samples were made out of difficult-to-cut materials (titanium based alloys – Ti6Al4V and nickel based alloys – Inconel 718). Milling process was carried out under different cutting parameters (cutting speed, feed and depth of cut). In order to check the possibility of neural networks for forecasting of residual stresses in case of machining of difficult-to-cut materials obtained result were implemented in neural networks.