Brushless DC Generator controlled by Constrained Predictive Algorithm

The brushless DC generator controlled by a predictive algorithm is considered in this paper. It is able to develop excellent performances such as minimum Joule losses and minimum torque ripple, at the same time. The tracking characteristic of the prime-mover is mandatory for setting the reference value of the electromagnetic torque developed by the generator, by means of which the switching pattern of the AC/DC converter is determined at each sampling time interval. The above generator performances are possible under certain constrained values of reference torque and rotor speed, due to the DC-bus voltage saturation. The knowledge of these quantities are necessary for the best matching of the prime-mover with the brushless DC generator and the AC/DC converter. In this paper, these constraints are investigated in detail with the aim of highlighting the best operation of the conversion system under a constant DC bus voltage.