Yenilik temelli uyarlanabilir genişletilmiş kalman filtresi kullanılarak asenkron motorun stator direnci kestirimi

In the industry, induction motors are widely utilized due to their low cost and low maintenance requirements. To get better performance from the motor, vector control (FOC) is introduced in the last decades. In this thesis, the software and hardware parameters that affect the performance of the vector-controlled induction motor will be investigated by simulations and experiments. PI controller design, axis decoupling, dc-link compensation and stator resistor estimation are implemented and simulated in Simulink then verified in an experimental setup. Having an extra sensor to sense the rotor speed brings an additional cost to the system; thus, sensorless vector control is widely used to control induction motors. Sensorless vector control generally uses the induction motor's voltage model to estimate the rotor flux; to accurately estimate the flux, the information of stator resistance value is crucial. An Innovation based adaptive extended Kalman filter model is implemented in the real-time environment to estimate the stator resistance of the motor. The experimental setup is controlled by DSPACE DS1104, which is programmed by the Simulink coder. The proposed algorithms are implemented and tested in Simulink first; after verification in the simulation environment, the algorithm is embedded into Dspace DS1104, and experiments are conducted.