Optimizing Machining Parameters in Production Using an Artificial Neural Network

Abstract

Setting the values for three controllable factors in metal cutting, namely feed, cutting speed and depth of cut, is considered as an important decision. The choice of the numerical values of the factors has a direct influence on productivity and costs of a machining operation. In this work, the response of a longitudinal turning is studied with different levels for the factors. Data collection from production would be valuable when optimizing machining parameters. Data collection in production is considered shortly and turning tests are made to study tool wear and tool life. Artificial Neural Network is used as a regression technique to model tool life. Two classical tool life equations are compared to the Artificial Neural Network. Simple turning simulation is made to study maximum production rate and minimum cost.