Auto-tuning of motor drive system by simple adaptive control approach

This paper is concerned with a new auto-tuning method to aci:lust the parameters in the feedforward and feedback PI controllers for a motor drive system, according to identified physical parameters such as a motor time constant, viscous friction terms proportional to motor velocity and squared velocity, and Coulomb friction. These physical parameters can be obtained by specially setting up a suitable reference model and plug-in models in the simple adaptive control (SAC) algorithm. Digital implementation for the proposed adaptive algorithm is also presented, in which the relative degree problem is solved by a serial compensation scheme for a non-ASPR system. The effectiveness of the proposed auto-tuning algorithm is examined in experimental studies in velocity control of an AC synchronous servo motor. It is clarified that the proposed scheme can simultaneously give the perfect tracking to the reference output and the specified attenuation performance of the closed-loop system, even if the load mass largely changes uncertainly.