Prediction of cutting forces using MRA, GMDH and ANN techniques in micro end milling of titanium alloy

Abstract

Micro End Milling is an specific procedure for manufacturing biomedical, automotive, naval, aerospace and mining parts especially for super alloys. Surface finish is the necessity output parameter after machining these alloys in the said applications. The study of cutting forces is mandatory to know the required surface finish to be generated on the work material. The influence of machining parameters can be investigated by experimental and predicted results that are very useful for the practitioner engineers inorder to evaluate. In this manuscript, experimental investigation of cutting forces was done and a comparison of the modelling methods - multiple regression analysis (MRA), Group method data handling (GMDH) and Artificial Neural Network (ANN) were presented. In the manuscript, the impact of input factors in machining of Ti-6Al-4 V (Grade-5) titanium alloy using uncoated tungsten carbide tools was studied on cutting forces. From the experimental interpretations, it is found that while increasing machining input parameters cutting forces on workpiece subsequently increases. After the adherences of experiments, experimental values are in good agreement with the accuracy of neural network prediction.