Browsing by Author "Zietsman, Nikita Linda"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Optimal design methodology of a three phase Rotary transformer for a brushless doubly fed induction generator
    (Stellenbosch : Stellenbosch University, 2016-03) Zietsman, Nikita Linda; Gule, N.; Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering.
    ENGLISH ABSTRACT: The brushless doubly fed induction generator (BDFIG) offers higher reliability over the standard doubly fed induction generator (DFIG), by eliminating the slip ring and brush assembly. The currently employed BDFIG configuration results in a more complex machine. In order to improve reliability while still retaining the simplicity of the DFIG, alternative brushless topologies have been explored. One such solution is to make use of a rotary transformer (RT) to replace the slip rings and brushes. The RT is placed on the same shaft as the generator and provides inductive coupling over the interface. Size and efficiency are important constraints of the RT used in a DFIG application. The purpose of this study is to develop a design methodology that takes into account the unique properties of the RT, while simultaneously minimising size and maximising efficiency. An analytical model, which makes provision for the non-idealities of the RT, is used to improve the analysis. The model is then verified using FEM. Although the improved model offers better correlation to the simulated results than the original model, there is still room for improvement. A prototype RT, is designed using developed procedure, constructed and practically tested. The resultant size is suitably smaller than a prototype built using standard transformer design strategies. Encouraging results are obtained from the tests, indicating that the RT performs well at the rated conditions. However, a number of differences exist between the FEM analysis and the measured operation. The cause of this deviation is discussed and suitable recommendations are made for future work.