A novel protection scheme for doubly fed induction generator‐based wind turbines during the fault occurrence in the rotor‐circuit

Abstract The rotor‐circuit (RC) of the doubly fed induction generator‐based wind turbine (DFIG‐WT) consists of various equipment in which faulty conditions must be immediately identified, because this faulty status leads to the loss‐of‐excitation (LOE) phenomenon and may cause some serious damages....

Full description

Bibliographic Details
Main Authors: Ali Rostami, Navid Rezaei
Format: Article
Language:English
Published: Wiley 2023-09-01
Series:IET Generation, Transmission & Distribution
Subjects:
Online Access:https://doi.org/10.1049/gtd2.12954
Description
Summary:Abstract The rotor‐circuit (RC) of the doubly fed induction generator‐based wind turbine (DFIG‐WT) consists of various equipment in which faulty conditions must be immediately identified, because this faulty status leads to the loss‐of‐excitation (LOE) phenomenon and may cause some serious damages. In this regard, this paper develops a novel solution to protect the DFIG‐WT during such a condition. For this purpose, a comprehensive investigation on behavior of the DFIG‐WT during the conditions of the RC failures has been carried out. The results show that, the output active power of the machine (Pdfig) can be considered as a suitable initial indicator for detecting the RC failures. However, to reliable detection of the partial‐LOE (PLOE), complete‐LOE (CLOE) and discrimination from the other power system disturbance (PSD), a directional over‐current (DOC) and an under voltage (UV) relays are suggested to be installed in the RC, at both ends of the back‐to‐back converters. Consequently, by detecting the Pdfig decrement, the provided DOC and UV relays in the RC will be triggered to classify the occurred events. The outlined protection scheme is exhaustively investigated on a typical power system which the obtained simulation results show the high reliability in discriminating the RC failures and PSD events with an acceptable detection time.
ISSN:1751-8687
1751-8695