Dissertation
ON-LINE PARAMETER IDENTIFICATION OF DOUBLY-FED INDUCTION GENERATORS FROM MEASUREMENT DATA
Doctor of Philosophy (PhD), Washington State University
01/2017
Handle:
https://hdl.handle.net/2376/12989
Abstract
Doubly-Fed Induction Generators (DFIG) are widely used nowadays for renewable energy generation. Control techniques need to be developed to match the growing DFIG wind turbine sizes and capacity. This requires better modeling verification tools of the DFIG which also bring other benefits. They can help to better increase their reliability, as well as improving the efficiency of the power output. This research mainly demonstrates the verification of DFIG electric models by identifying the parameters of operating DFIGs. The 4th order and 2nd order DFIG electric models were derived and specifically used for the parameter identification process. In order to provide the measurement input data for the parameter identification algorithm, a customized DFIG simulation model was built in MATLAB SIMULINK. By applying the Model Reference Adaptive Control (MRAC) algorithm, the electric parameters of the DFIG models have been identified with high accuracy and short convergence time. The proposed MRAC algorithm also performs well on identifying the parameters when the rotational speed changes. In the field, the measurement data can be taken from Phasor Measurement Units (PMU) to achieve the on-line identification. This identification technique can be used as an ad-hoc function block as well, that can cooperate with other DFIG control techniques, as well as provide reference information indicating the real-time operation condition of the wind- turbines.
Metrics
25 File views/ downloads
32 Record Views
Details
- Title
- ON-LINE PARAMETER IDENTIFICATION OF DOUBLY-FED INDUCTION GENERATORS FROM MEASUREMENT DATA
- Creators
- SHAOTONG GUO
- Contributors
- Anjan Bose (Advisor)Chen Ching Liu (Committee Member)Ali Mehrizi Sani (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- School of Electrical Engineering and Computer Science
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 107
- Identifiers
- 99900581719201842
- Language
- English
- Resource Type
- Dissertation