Title: Identification of machining force model parameters from acceleration measurements

Authors: Bartosz Powalka, Jaspreet S. Dhupia, A. Galip Ulsoy, Reuven Katz

Addresses: Szczecin University of Technology, Piastow 19, 70-310 Szczecin, Poland. ' Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109-2125, USA. ' Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109-2125, USA. ' Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109-2125, USA

Abstract: This paper presents a method, which does not require the use of a force dynamometer, for estimating the mechanistic cutting force model coefficients in milling. The cutting forces are reconstructed from acceleration signals measured during machining, and from experimentally evaluated Frequency Response Functions (FRFs). The quality of reconstruction is improved by the application of the Tikhonov regularisation technique, for which the measurement noise is estimated beforehand. Since the DC component of the FRF and of the acceleration signals is unknown when using piezoelectric accelerometers and impact hammers, the cutting force model estimation requires the knowledge of the cutting force pattern for a particular cutting scenario.

Keywords: cutting forces; regularisation; system identification; stability lobes; milling; cutting force models; acceleration measurements; acceleration signals; machining; frequency response; cutting force patterns.

DOI: 10.1504/IJMR.2008.019210

International Journal of Manufacturing Research, 2008 Vol.3 No.3, pp.265 - 284

Published online: 02 Jul 2008 *

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