Implementation of Predictive Controllers for Matrix-Converter-Based Interior Permanent Magnet Synchronous Motor Position Control Systems

Research output: Contribution to journalArticlepeer-review

22 Citations (Scopus)

Abstract

This paper proposes the implementation of predictive controllers for matrix-converter-based interior permanent magnet synchronous motor (IPMSM) position control systems. A model-based predictive position controller and a model-free predictive current controller (MFPCC) for IPMSMs are investigated here. The proposed predictive controllers can improve the dynamic responses, including transient, load-disturbance, and tracking responses. In addition, the proposed MFPCC does not require any parameters such as inductance, resistance, or back-electromotive force of the IPMSM, and its performance does not deteriorate because of variations in the virtual dc-link voltage. Only the stator current difference is used to predict the future sampling current. A detailed stability analysis of the current-loop control and position-loop control of the IPMSM position control system is discussed. Several experimental results are included to validate the theoretical analysis. The experimental results show that the proposed predictive controllers have better performance than PI controllers. A 32-bit digital signal processor, a TMS-320LF-2407A, is used to execute the predictive controllers. The proposed control system is easily applied in industry due to its systematic design procedure.

Original languageEnglish
Article number8478353
Pages (from-to)261-273
Number of pages13
JournalIEEE Journal of Emerging and Selected Topics in Power Electronics
Volume7
Issue number1
DOIs
Publication statusPublished - Mar 2019
Externally publishedYes

Keywords

  • Interior permanent magnet synchronous motor (IPMSM)
  • matrix converter
  • predictive controller

Fingerprint

Dive into the research topics of 'Implementation of Predictive Controllers for Matrix-Converter-Based Interior Permanent Magnet Synchronous Motor Position Control Systems'. Together they form a unique fingerprint.

Cite this