Title: On-line monitoring of surface roughness in turning operations with opto-electrical transducer

Authors: Avisekh Banerjee, Evgueni V. Bordatchev, Sounak Kumar Choudhury

Addresses: Department of Mechanical and Materials Engineering, University of Western Ontario, London, Ontario, Canada. ' National Research Council of Canada, Industrial Materials Institute, London, Ontario, Canada. ' Department of Mechanical Engineering, Indian Institute of Technology, Kanpur, India

Abstract: This work studies the feasibility of on-line monitoring of surface roughness in turning operations using a developed opto-electrical transducer. Regression and Neural Network (NN) models are exploited to predict surface roughness and compared to actual and on-line measurements. The comparative study suggests feasibility of using the transducer within 15% tolerance. Pattern recognition analysis of on-line roughness and vibration displacements is used for reliable (>93%) classification of actual roughness. The results provide important information for the future development of on-line diagnostics and control of surface roughness in turning operation. [Received 4 January 2008; Revised 14 April 2008; Accepted 9 June 2008]

Keywords: online monitoring; turning operations; surface roughness; bifurcated optoelectrical transducers; regression models; neural networks; modelling; pattern recognition; vibration displacement.

DOI: 10.1504/IJMR.2009.022743

International Journal of Manufacturing Research, 2009 Vol.4 No.1, pp.57 - 73

Published online: 25 Jan 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article