State Space Modeling Theory of Induction Motors for Sensorless Control and Motoring Purposes

Abstract

The purpose of this work is to investigate different models of the asynchronous machine in the state space representation for sensorless control purposes. The asynchronous machine is known as a complex non-linear system in which time–varying parameters entail an additional difficulty. Based on the fact that model can be significantly simplified if one applies the d-q Park transformation and that high dynamic performances are achieved by a field orientation technique also called vector control, different structures of the model are investigated and discussed. Each model merits and drawbacks are pointed out.

This work contributes to decrease of the number of sensors to be used at the machine without deteriorating the dynamic performance of the considered drive control system. With the development of parameter and state estimation techniques, mainly the observer technique such as Kalman filtering and Luenberger observer, one can, based on the available measurements, estimate the remaining parameters or state components. Thus allows us to increase the reliability and the robustness of drive systems and to decrease their cost.

Based on this point of view state feedback control of a linear state space model of the asynchronous machine involving the stator current and flux as state variables is experimented in a simulation study. The sensorless state feedback control uses the estimated stator flux and the measured stator current. The simulation permits also to estimated the remaining state variables of the asynchronous machine such as rotor current and flux.

The obtained results show that the state space technique is an efficient method for asynchronous machine modelling, state variables estimating and sensorless machine drives controlling. State space techniques are directly used by the observer-based as well as the Kalman filtering techniques for sensorless control and monitoring purposes.