Adaptive Machine Tool System Regulation

Abstract

Least effort controller design procedures, for multivariable, machine tool models, are considered. The employment of these regulators for adaptive system analysis is proposed. An approximate relationship between the nonlinear, stochastic process system dynamics and the linear, mid-range model is employed. Model-reference, adaptive, multivariable control is advocated with the model–system error vector augmenting the actuator signals, during the system transient. The amplitude of the model-system uncertainty and the guarantee of closed-loop stability are established as mandatory performance requirements. To illustrate the theoretical procedures outlined, a nonlinear, distributed-lumped parameter model is employed, in a machine tool system application study. An adaptive model-reference control strategy is formulated using a simple, analytical representation for the system. The response of the adaptive system, following input and cutting load disturbances, is computed and the effectiveness of this form of controller is commented upon.