Maximum power point tracking and sensorless control for permanent magnet generators tested on a wind turbine emulator

This paper presents a maximum power point tracking algorithm (MPPT) and sensorless control for permanent magnet generators associated with small scale wind turbines. The basis for the wind energy conversion is presented to model the wind turbine emulator test-rig and support the proposed MPPT algorithm. The maximum power point tracking algorithm is based on estimations of the load torque and shaft speed. The shaft speed and position are estimated with an extended Kalman filter. The load torque is estimated with a discrete time observer. A sliding mode controller in the rotor reference frame drives the generator in speed control mode. A minimum number of sensors is required for this implementation: one voltage sensor for the DC link and two current sensors for the phase currents. Experimental implementation on the wind emulator test-rig is given. The proposed estimators, sensorless controller and MPPT algorithm were shown to be well integrated and work properly. The sensorless controller was able to track a speed reference calculated by the MPPT algorithm and based on the speed and load torque estimations.