Dual-Subspace Model Predictive Control for Dual Three-Phase Permanent Magnet Synchronous Generators
摘要
Model predictive control has been widely applied in multiphase motor control systems due to its advantages of fast dynamic response and simple structure. However, traditional model predictive control primarily focuses on electromechanical energy conversion in the fundamental subspace, leading to unsatisfactory harmonic current suppression. This paper proposes a dual-subspace composite model predictive power control scheme for a dual three-phase permanent magnet synchronous generator system, by fully considering its multi-vector and multi-degree-of-freedom characteristics. Firstly, virtual voltage vectors are synthesized in both the fundamental and harmonic subspace to achieve coordinated control of the two subspaces. On this basis, a composite control strategy is adopted, where model predictive power control is employed in the fundamental subspace and current deadbeat predictive control in the harmonic subspace. Combined with a duty cycle modulation strategy, the steady-state control performance can be significantly improved. Experimental results verify the feasibility and effectiveness of the proposed control strategy.