Sensorless induction machine observation with nonlinear magnetic characteristic
Résumé
This paper addresses the problem of state estimation in induction motors. Generally, the motor observer design has been dealt based on standard models neglecting the saturation effect in the magnetic characteristic. As a matter of fact, magnetic saturation cannot be ignored especially when considering control strategies (speed and torque) that involve large flux variations. Such large flux variations are necessary to meet optimal operation conditions in the presence of wide range load torque changes. On the other hand, it is well known that use of mechanical (speed and torque) sensors entails reliability issues. In this paper, a new sensorless adaptive observer is designed for induction machine based on a model that accounts for the nonlinear feature in the magnetic circuit. The observer consists of two interconnected state-dependent gain observers and is formally shown to provide accurate estimates of the mechanical and magnetic variables using only stator current and voltage measurements