Skip to main navigation Skip to search Skip to main content

Ultralocal-Model Generalized Integrator Observer Based Current Reconstruction Strategy for Single DC-Link Shunt PMSM Drives

  • School of Electrical Engineering and Automation, Harbin Institute of Technology
  • Imperial College London

Research output: Contribution to journalArticlepeer-review

Abstract

Due to the limitations of the single current shunt sampling topology and the minimum sampling time of the permanent magnet synchronous motor drive system, there is a current sampling dead region (CSDR) at the operation of high modulation index (MI). In order to expand the output voltage range of the inverter and improve the parameter robustness, a phase current reconstruction (PCR) strategy with ultralocal-model-based generalized integral current observer is proposed. The impact of CSDR on the inverter output voltage is analyzed, and the parameter sensitivity of the PCR strategy with the model-based observer is clarified. Based on the constructed ultralocal-model, a generalized integral current observer is designed to realize PCR at the operation of high MI. The stability of the proposed generalized integral current observer is analyzed, and the influence of its characteristic parameters is explained. The proposed strategy has higher reconstruction accuracy and better parameter robustness. The effectiveness of the proposed strategy was verified on the 2.2-kW interior permanent magnet synchronous compressor drive platform.

Original languageEnglish
Pages (from-to)10758-10768
Number of pages11
JournalIEEE Transactions on Power Electronics
Volume40
Issue number8
DOIs
StatePublished - 2025
Externally publishedYes

Keywords

  • Current observer
  • permanent magnet synchronous motor (PMSM)
  • phase current reconstruction (PCR)
  • single current shunt sampling
  • ultralocal-model

Fingerprint

Dive into the research topics of 'Ultralocal-Model Generalized Integrator Observer Based Current Reconstruction Strategy for Single DC-Link Shunt PMSM Drives'. Together they form a unique fingerprint.

Cite this