Loading...

Modeling and Data Mining of Partial Discharge in Power Transformer Solid Insulation

Jahangir, Hamid | 2012

542 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 43650 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Vakilian, Mehdi
  7. Abstract:
  8. Transformers are one of the most important equipments in transmission and distribution networks. Transformer unplanned outages have severe impacts on the continuity of power system operation. To improve the reliability of transformers and to achieve an optimum operation cost, online condition monitoring of transformers is inevitable. Information about the quality of the transformers insulation system is known as the best parameter to be monitored in a transformer. Since partial discharge signals are initiated long before the beginning of a severe damage, partial discharge monitoring and its evaluation canbe employed to warn the operator.Data mining on the partial discharge signals extracts useful information from large volumes of measured data. Due to presence of anoisy environment, the recorded signals areembedded in the excessive noise pulses. Thus at the first step, the measured partial discharge need to be de-noised. Once the denoising process is completed,the partial discharge sources are identified and classified.In this thesis, the physical models of transformer solid insulation defectsare constructed in laboratory and its partial discharge signals are measured. Thena noise removal method is employedto detect partial discharge pulses in presence of severe environment noise.In the final stage, classification of partial discharge sources using data mining techniques in time and phase domain is presented.
  9. Keywords:
  10. Power Transformer ; Partial Discharge ; Data Mining ; Denoising ; Transformer Solid Insulation

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC