Title :
New statistical methods for evaluation of DGA data
Author :
Mosinski, F. ; Piotrowski, T.
Author_Institution :
Dept. of High Voltage, Tech. Univ. Lodz, Poland
fDate :
4/1/2003 12:00:00 AM
Abstract :
This paper presents a statistical method for generation of the signals warning of a possibility of failure in oil filled power transformers. The method uses the dissolved gas analysis (DGA) data to produce quantitative values warning of the beginning and progress of failure. Statistical tests are employed to test the bimodality of four standard distributions: normal, log-normal, Weibull and Gumbel. Numerical examples are also presented.
Keywords :
Weibull distribution; chemical analysis; insulation testing; log normal distribution; normal distribution; power transformer insulation; transformer oil; EHV transformers; Gumbel distribution; UHV transformers; Weibull distribution; bimodality; dissolved gas analysis; failure; log-normal distribution; normal distribution; oil filled power transformers; quantitative values; signals generation; statistical methods; Dissolved gas analysis; Oil insulation; Petroleum; Power generation; Power transformer insulation; Power transformers; Statistical analysis; Statistical distributions; Testing; Weibull distribution;
Journal_Title :
Dielectrics and Electrical Insulation, IEEE Transactions on
DOI :
10.1109/TDEI.2003.1194108
