A class of flux observers for doubly-fed induction generators used in small power wind generation systems

C. Lascu, I. Boldea, Frede Blaabjerg, W. Chen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

8 Citations (Scopus)

Abstract

This paper investigates a family of stator and rotor flux observers for sensorless operation of doubly-fed induction generators (DFIG). Four stator flux observer topologies are described and compared. All proposed schemes use the voltage and current models connected in parallel or in series. For each scheme, a proportional-integral (PI) compensator is used to form a feedback correction loop that encloses one or both models. The estimation errors of either the stator current, rotor current, stator flux, or rotor flux are used as inputs for the PI blocks. The speed estimation is based on a model reference adaptive system and a PLL observer. The reference quantity is the measured rotor current, the adaptive model is a current estimator, and the adaptive quantity is the rotor position. Comparative experimental results with all observers on a 7.5 kW DFIG system are presented and discussed in the paper. It is concluded that the observers are able to produce accurate results and are suitable for practical applications.
Original languageEnglish
Title of host publicationProceedings of the IEEE Energy Conversion Congress and Exposition, ECCE 2013
Number of pages7
PublisherIEEE Press
Publication date2013
Pages2289-2295
ISBN (Print)978-1-4799-0336-8
ISBN (Electronic)978-147990335-1
DOIs
Publication statusPublished - 2013
EventIEEE Energy Conversion Congress and Exposition - Denver, United States
Duration: 15 Sep 201319 Sep 2013

Conference

ConferenceIEEE Energy Conversion Congress and Exposition
CountryUnited States
CityDenver
Period15/09/201319/09/2013

Fingerprint Dive into the research topics of 'A class of flux observers for doubly-fed induction generators used in small power wind generation systems'. Together they form a unique fingerprint.

Cite this