NONLINEAR ADAPTIVE MODELING FOR MACHINING CHATTER IDENTIFICATION AND MONITORING.

This paper presents a nonlinear adaptive modeling scheme for the identification and monitoring of turning processes on a NC lathe machine. In the application of the adaptive model to the analysis of machine tool chatter vibration, the entire NC machine-tool system, is treated as a black box. The input cutting forces are assumed to be a stochastic function of chip thickness, feed rate and various random disturbances. The onset of the chatter vibration is characterized by an adequate nonlinear autoregressive (NLAR) model obtained directly from the vibration response data. The monitoring of the process is made by the employment of a nonlinear adaptive forecasting control chart (NAFCC) scheme.