A Delayed State Kalman Filter for On-line Estimation of Induction Motor Parameters and Rotor Flux Space Vector Position

This paper deals with a new delayed state Kalman filter to be used to simultaneously estimate all the induction motor parameters and the rotor flux space vector position that are necessary to correctly implement direct field-oriented control schemes. The algorithm is based on the extended Kalman filter theory and only uses machine quantities that can be easily measured: the stator voltages and currents, and rotor speed. The on-line adaptation of all the model parameters by means of the estimated ones makes it possible to obtain very accurate estimates of the rotor flux space vector position, in spite of temperature and magnetic saturation variations. This is the main advantage of the proposed filter. It is believed that it is superior to the other ones, based on Kalman filter theory, that estimate only the rotor time constant. The results of the application of the new algorithm fully prove its effectiveness.