Predictive Duty Cycle Control of Three-Phase Active-Front-End Rectifiers

This paper proposed an on-line optimizing duty cycle control approach for three-phase active-front-end rectifiers, aiming to obtain the optimal control actions under different operating conditions. Similar to finite control set model predictive control strategy, a cost function previously constructed based on the desired control performance is adopted here, which is essential for the solving process of the optimizing problem. On the other hand, differently, with respect to the proposed strategy, duty cycle signals are optimized, instead of possible switching states. The determination of optimal duty cycles is made by predicting the effect of duty cycles on instantaneous current variations and minimizing the cost function. Due to the adoption of behavior prediction, the proposed controller inherits the excellent dynamic characteristics of predictive controllers. Moreover, the application of optimal duty cycles determined by cost function minimization automatically ensures optimum operations of converters within each sampling period. Improved transient and steady-state features of the proposed strategy are confirmed by experimental validations and in-depth comparisons with linear controllers, deadbeat predictive controllers, and finite control set model predictive control strategies.