Prediction of surface roughness by measuring flank wear and cutting forces in turning process

K. Deepak Lawrence, L. M. Lakshmanan, S. Sathish Kumar

Abstract

Modeling and prediction of surface finish of work pieces in machining can play an important role in the automation of manufacturing operations. As surface generation in machining is a complex process, the exact variables and parameters that have to be used for surface roughness prediction models are still under dispute. This work attempts to evaluate and predict surface roughness during turning operation using artificial neural network and multiple regression analysis. In addition to the conventional cutting conditions like cutting speed, feed and depth of cut, this work also used cutting force ratio, cutting time and flank wear of the cutting tool to train the predictive models. The developed models were found to be capable of better predicting the surface roughness with very minimum RMS error and high correlation coefficient compared to earlier works.

Keywords

Surface Roughness - Flank wear - Back Propagation Neural Network – Multiple Regression Analysis

Download

IMEKO-TC14-2007-11.pdf

DOI

-

IMEKO TC

TC14 - Measurement of Geometrical Quantities

Back to the proceedings