Passivity-based Rotor Time Constant Estimation and Adaptive Position Control of Induction Machines

This paper presents an adaptive controller based on the passivity theorem to control the position of an induction machine. Firstly, an open-loop flux observer estimates the rotor flux linkages, and then adaptive laws are designed to estimate the inverse rotor time constant, mechanical parameters, and load torque. The passive properties of the feedback connection of the rotor flux observer to the inverse rotor time constant estimator, and the adaptive controller are analyzed by the passivity theorem. The overall control system is formally verified to be globally stable by the theorem instead of Lyapunov-like functions. Experimental results show that the parameter estimation errors converge to zero when the position command is a sinusoidal signal with two distinct frequencies. Furthermore, good position tracking responses can be obtained, and the approach is robust with regard to the variation of the parameters and load torque disturbances.