Surface qualify prediction by artificial-neural-networks

Surface roughness is often taken as an important indicator of the quality of machined parts. Achieving the desired surface quality is of great importance for the product function. In this paper, influence of material, type of tool, cutting depth, feed rate and cutting speed on surface roughness is observed. Collected results of experimental research are utilized for surface roughness prediction using neural networks. Various structures of a back-propagation neural network have been analyzed and the network with the minimum RMS error has been selected. Evaluation of surface roughness obtained by neural networks model can help to less experienced technologists and therefore production preparation technological time can be shorter.