Hadi Fateh - Phd Student, Islamic Azad University, Sanandaj Branch, Sanandaj, Iran, Iran
Rahim Shamsi Varzeghan - MSc at Electrical Engineering, Azarbaijan Shahid Madani University, Tabriz, Iran
Mohammad Reza Jannati Oskuee - Phd Student, Electrical Engineering Department, Azarbaijan Shahid Madani University, Tabriz, Iran

Inter-turn faults have a significant role in power transformers and neglecting them can result more serious damages in power transformers. However, there is not a great study on this subject so far. Sweep Frequency Response Analysis (SFRA), based on comparisons between frequency responses, is a powerful method to detect these faults. Analyzing of the frequency response to detect the transformer faults requires the experts to interpret the test results. The main problem with SFRA is to need an expert view in failure detecting. The most accurate solution for this problems using statistical indicator. This paper means to propose a number of nonparametric statistical methods that have not been used so far. Then these equations are applied to some different sets of experimental SFRA measurements. Finally, some nonparametric statistical indicators that have good treat in recognizing inter-turn faults, using SFRA measurement, are presented.

Power transformers, Frequency response analysis, Nonparametric statistical methods, Inter-turn fault, FRA interpretation