• DocumentCode
    3158337
  • Title

    Make use of UV/VIS spectrophotometer to determination of dissolved decay products in mineral insulating oils for transformer remnant life estimation with ANN

  • Author

    Malik, Hasmat ; Yadav, Amit Kr ; Tarkeshwar ; Jarial, R.K.

  • Author_Institution
    Electr. Eng. Dept., Nat. Inst. of Technol., Hamirpur, India
  • fYear
    2011
  • fDate
    16-18 Dec. 2011
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    UV-Spectrophotometer response is a nonintrusive test used to determine the transformer integrity. Information about the health of the power transformer that can be use to plan cost, maintenance, relocation and operational criteria can be accurately interpreted using UV-Spectrophotometer. As UV scan can only show the pictorial information of the age of the oil hence it is not advantageous in all aspects. In the present paper an Artificial Neural Network (ANN) method to determine the health assessment of the transformer oil introduced. The ANN uses the UV/Vis spectroscopy absorbance and bandwidth values of the transformer oil which are in service at several locations. The ANN method is designed so that the dissolved decay product in mineral insulating oil of transformer can be examined quickly and automatically.
  • Keywords
    minerals; neural nets; power engineering computing; power transformer insulation; transformer oil; ultraviolet spectroscopy; ANN; UV-VIS spectrophotometer response; artificial neural network method; dissolved decay products determination; mineral insulating oils; nonintrusive test; power transformer remnant life estimation; Aging; Artificial neural networks; Minerals; Oil insulation; Power transformer insulation; Training; Aging; Ultraviolet spectrophotometer; artificial neural network (ANN); mineral oil; transformer;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    India Conference (INDICON), 2011 Annual IEEE
  • Conference_Location
    Hyderabad
  • Print_ISBN
    978-1-4577-1110-7
  • Type

    conf

  • DOI
    10.1109/INDCON.2011.6139574
  • Filename
    6139574