Controle de velocidade sensorless de motores brushless DC submetidos a variações periódicas de carga

Abstract

This master thesis presents a sensorless speed control drive of Brushless DC motors used in reciprocating compressors. These compressors cause periodic load torque variations on the motor, consequently mechanical vibrations and acoustic noise are presented as well. As a result, a discrete speed controller based on the internal model principle is proposed. This controller aims to track the speed reference and to reduce the periodic torque disturbances caused by the compressor. The proposed controller consists of a rotor position and speed estimator, and a discrete speed controller with variable sampling frequency that depend on the rotor speed. The discrete estimator is based on the zero crossing point detection of the back-electromotive force. This method is optimized by using a Least Squares algorithm. The speed controller consists of two control actions, repetitive and proportional-derivative actions. In order to complement the drive system operation, a new procedure to start a BLDCM with current control is proposed. This technique ensures the maximization of the electromagnetic torque during startup and to avoid demagnetization of the permanent magnet by the stator current. Finally, simulation and experimental results are presented to demonstrate the performance of the techniques adopted, as well as the main advantages, disadvantages and operational limitations of the experimental implementation. In addition, the performance of the load torque rejection is evaluated by comparing the harmonic spectrum of the torques produced by the proposed speed controller and a PI controller.