Skip to main navigation Skip to search Skip to main content

A Novel SOGI-PLL Incorporating Adaptive Synchronous Frequency Extraction Filter for SPMSM High-Precision Position Estimation

  • School of Electrical Engineering and Automation, Harbin Institute of Technology

Research output: Contribution to journalConference articlepeer-review

Abstract

High-frequency signal injection sensorless algorithms have been widely studied and used for rotor position estimation in permanent magnet synchronous motors (PMSMs) at low speeds or stationary states. The traditional position sensorless control method results in harmonic fluctuations in position estimation owe to the 5th and 7th current harmonics brought by the inverter nonlinearity. In order to minimize this problem, this paper proposes an improved phase-locked loop (PLL) position estimation method based on high-frequency square-wave signal injection to suppress position error fluctuations, which is applied in the zero-low-speed domain of surface-mounted motors. Two adaptive synchronous frequency extraction filters (ASFEF) and a second-order generalized integrator (SOGI) are introduced into the conventional phase-locked loops to extract the fundamental waveform for rotor position estimation and adaptively eliminate the 6th harmonic component of position estimation error in the feedforward channel. The comparative experimental results verify the validity and feasibility of the proposed method.

Original languageEnglish
JournalSymposium on Sensorless Control for Electrical Drives, SLED
Issue number2025
DOIs
StatePublished - 2025
Externally publishedYes
Event12th IEEE International Symposium on Sensorless Control for Electrical Drives, SLED 2025 - Harbin, China
Duration: 15 Aug 202517 Aug 2025

Keywords

  • Permanent magnet synchronous motor (PMSM)
  • Sensorless control
  • phase-locked loop (PLL)
  • position estimation

Fingerprint

Dive into the research topics of 'A Novel SOGI-PLL Incorporating Adaptive Synchronous Frequency Extraction Filter for SPMSM High-Precision Position Estimation'. Together they form a unique fingerprint.

Cite this