A Novel Three-Vector-Based Model Predictive Flux Control With Low Computation Complexity for SPMSM

Abstract

A novel three-vector-based model predictive flux control (TV-MPFC) with low computation complexity and low stator flux ripple for SPMSM is proposed, in which the stator flux vector tracking error is evaluated in the cost function, so the weighting factor used in model predictive torque control (MPTC) is eliminated, thus avoiding tedious tuning work. To enhance the steady-state performance, three voltage vectors including two active vectors and one zero vector are applied in one period, and the key improvement of the proposed strategy is to construct a novel voltage vector selection table for the second optimal active vector based on the stator flux vector error which decreases the number of candidate active vectors, thus reducing the computation complexity. The single-vector-based MPFC and four three-vector-based model predictive control algorithms are compared in this paper. The comparative experimental tests are finally implemented and the results show that the proposed TV-MPFC can enhance steady-state performance of stator flux while reducing computation complexity, hence it is more suitable for scenarios with higher requirement on stator flux.