Prediction of roughness of planar surfaces processed with toroidal milling through an artificial neural network

Abstract

This paper is intended to create an artificial neural network capable of generating new values for the roughness on the basis of experimentally obtained data bases. Experimentally you will measure the roughness of the flat surfaces processed with the toroidal milling, the process factors being the input neurons of the neural network, following the roughness values being the output neurons. It aims to modify the input neurons from the same neural network and generate new roughness values.