Matrix converter is an AC-AC direct power converter comprising of an array of bi-directional switches. It does not require an intermediate DC-link and allows sinusoidal output waveforms with varying amplitudes and frequencies. The configuration of these bi-directional switches decides the number of inputs and outputs of the matrix converter. This research uses a direct matrix converter (DMC) as a phase-changing device that can convert a three-phase AC voltage into a 5-phase AC voltage. The DMC is modulated with the model predictive control algorithm. The output of DMC is fed to a five-phase permanent magnet synchronous motor (PMSM). The model predictive current control technique for DMC is carried out by developing a mathematical model of an input filter and PM motor used as a load. The predictive control of DMC results in sinusoidal output current, and it also enables the frequency variation in the output current. This frequency variation is useful in controlling the speed of the motor connected to the load. After controlling the 5-phase motor, the switching frequency regulation is done to observe its effect on the motor's stator current waveforms. Switching frequency regulation helps to limit the unnecessary switching of DMC. We developed a MATLAB-based Simulink model to study PMSM, and detailed results are presented. The results show that switching regulation can significantly reduce the switching frequency without compromising the current waveform quality.
Research on Model Predictive Control Based on Sensorless Permanent Magnet Synchronous Motor
