Model predictive power control of dual-stator brushless doubly-fed induction generator with reduced power ripple

The dual-stator brushless doubly-fed induction generator (DS-BDFIG), with its low-cost nature, high reliability operation and high torque/power density, shows great potential in variable speed generation applications. This paper explores the power quality improvement for grid-connected operation of the DS-BDFIG. An improved model predictive power control (MPPC) is proposed to reduce power ripple and lower current THD. A zero vector is applied along with an active vector in each control period. The duration of the selected active vector is optimized by power error minimization principle. Besides, the active vector selection and its duration are simultaneously considered in evaluating the cost function to ensure global minimization. The superiority of the proposed MPPC in power ripple reduction has been validated by a series of comparative simulations.