Prof. Keshav H. Jathkar

Dr. S. S. Mantha

Dr. Arun Kumar

A.A. Qazi

Abstract

In this paper an attempt has been made on analysis of BTA machining using neural network, which has not been done so far. Boring Trepanning Association (BTA) machining are capable of drilling holes having large length to diameter ratio in a single pass. The present work deals with surface roughness and hole size of BTA machined stress proofsteel components using back propagation neural network (BPNN). Levenberg - Marquardt algorithm is used to speed up the convergence of back propagation. Experiments have been carried out on production equipment over a wide range of cutting conditions and the effect of various process parameter like spindle speed,feed rate and length ofhole surface roughness and hole size has been studied. The data thus collected from the experiment has been used to train a BPNN for surface roughness and hole size. The performance of the trained network has been tested with the experimental data, andhas been found to be satisfactory.

Keywords- Artificial neural network, BTA machining, surface roughness, hole size.