Frequency Adaptive ADRC-Based Current Control for Grid-Connected Inverters in Weak Grids

  • Qiangsong Zhao
  • , Yuming Zhang
  • , Yuanqing Xia*
  • , Ji Feng Zhang
  • , Jun Wang
  • , Xu Fang
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Traditional active disturbance rejection control (T-ADRC) schemes for grid-connected inverters (GCIs) face challenges in reference tracking accuracy and harmonic disturbance rejection under grid frequency fluctuations, due to the limited observer bandwidth and phase lag issues. To solve these problems, this article proposes a frequency adaptive ADRC (FA-ADRC) scheme, integrating a frequency adaptive repetitive control-based extended state observer (FA-RC-ESO) and a frequency adaptive quasi-proportional-resonant (FA-QPR) controller. The FA-RC-ESO employs an infinite impulse response filter to enhance harmonic disturbance estimation, while the FA-QPR achieves zero-phase shift and unity gain for accurate reference tracking. In addition, the observer bandwidth is optimized by designing zero-phase low-pass filter in the internal model of repetitive control, and the sensor-induced high-frequency noise is effectively suppressed. Theoretical analysis confirms both stability and frequency adaptability of the proposed FA-ADRC. Experimental tests on a 1.5-kW single-phase GCI demonstrate superior performance: Precise reference tracking, improved harmonic suppression, and enhanced frequency adaptive capability.

Original languageEnglish
Pages (from-to)5053-5066
Number of pages14
JournalIEEE Transactions on Power Electronics
Volume41
Issue number4
DOIs
Publication statusPublished - 2026
Externally publishedYes

Keywords

  • Active disturbance rejection control (ADRC)
  • extended state observer (ESO)
  • frequency adaptability
  • grid-connected inverter (GCI)
  • repetitive control (RC)

Fingerprint

Dive into the research topics of 'Frequency Adaptive ADRC-Based Current Control for Grid-Connected Inverters in Weak Grids'. Together they form a unique fingerprint.

Cite this