Speed Sensorless Vector Control of an Induction Motor using Spiral Vector Model-ECKF and ANN Controller

Abstract

This paper presents a speed sensorless vector control of an induction motor using an extended complex Kalman filter, a neural network, a spiral vector model and two sensors for tracking voltage and current of one phase of stator. The spiral vector model uses the spiral vector variables rotating counter clockwise in the complex plane. This model depends only on variables and parameters of one phase of stator and one phase of rotor without Park transformation. The rotor speed, airgap flux and stator current of one phase are estimated by a new variant of the extended Kalman filter in the complex domain. The estimated rotor speed, airgap flux and stator current are used for vector control where all controllers are based on the neural network. Computer simulations have been carried out to test the effectiveness and robustness of the proposed control under noise and several load torques.