Lagrange Multipliers Aided Modulated Model Predictive Control Technique for PMSM Drives

Abstract

The modulated model predictive control (MMPC) method utilizes multiple voltage vectors in sequence within one switching cycle, achieving a fixed switching frequency. However, the control performance of using the MMPC technique still suffers from imprecise vector duration derivation and high computational burden. To address this, this paper proposes an improved MMPC technique aided by the Lagrange multiplier method to accurately calculate the duration of each vector, and then efficiently select the optimal voltage vector. Compared to the existing MMPC techniques, the proposed one exhibits a lower computational burden and superior steady-state performance. Those benefits have been proven by the experimental results of a 1.5 k rpm permanent magnet synchronous motor drive. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.