Advanced parameter identification for a linear-motor-driven motion system using disturbance observer

Abstract

Disturbance observer (DOB) is generally introduced into motion control systems to eliminate undesired disturbances and plant uncertainty. The DOB is also used for system identification. This work presents a novel experimental identification algorithm using disturbance observer to identify inertia, viscous coefficient, and friction of linear-motor-driven motion system. A conventionally adopted algorithm for determining the inertia of the motion system based on orthogonal relations among system responses is modified and extended to estimate the viscous coefficient and the magnitude of Coulomb friction of the underlying system. The advantages of the proposed method are high convergence rate and only one experiment needed to evaluate the system parameters. The proposed algorithm is demonstrated to be workable by both simulation and experiment.