A Simplified Identification Method for Autoregressive Models of Cutting Force Dynamics

Abstract

The quality and the duration of the metal removal process are two fundamental topics of today’s manufacturing technology. This paper presents a technique addressing the quality issue. There are numerous parameters involved in the dynamics of the cutting process. Some of them are treated as the major control parameters in this study, such as the depth of cut, feed rate, and the axial tool position. The tool-tip/workpiece interactions are elaborated, with the hope that a close correlation exists betwen the quality of the workpiece surface and the tool forces. The variations of the control parameters are examined in order to evaluate their influences on the tool forces. A numerical representation of the tool force dynamics is introduced using the time series analysis. This numerical technique is expanded to generate some relationships between the identifiers of the time series and the above stated control parameters. These relationships require a prohibitively large number of individual tests. The main contribution of the present study is to reduce the required number of tests by using this expanded numerical technique. Additionally, this technique leads itself to be utilized as an on-line identification scheme. This identification process is expected to comprise a portion of the future control strategies of metal machining. The experimental force registrations and the corresponding values obtained from the numerical representation are compared in an error analysis. Close agreement of the two groups of values is observed and tabulated within the text.