Novel Technique to Improve the Fault Detection Sensitivity in Transformer Impulse Test

• Published in: IEEE transactions on power delivery Vol. 23; no. 2; pp. 717 - 725
• Main Authors: Al-Ammar, E., Karady, G.G., Sim, H.J.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.04.2008; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Applied sciences; Dielectrics and electrical insulation; Electric potential; Electrical engineering. Electrical power engineering; Exact sciences and technology; Fast Fourier transforms; Fault detection; Faults; Fourier transforms; Impulse testing; Impulses; Insulation; Insulation testing; Performance evaluation; Power transformer insulation; signature analysis; Testing. Reliability. Quality control; time-frequency representation; Transfer functions; transformer; Transformers; Transformers and inductors; Voltage; Voltage-controlled oscillators; Electric transformer; Fourier transformation; Fast Fourier transformation; transformer; Frequency domain method; Electrical insulation; time-frequency representation; signature analysis; Faults; Sensitivity; Transfer function; Defect detection; Machine windings
• ISSN: 0885-8977; 1937-4208
• DOI: 10.1109/TPWRD.2008.915841

Insulation is identified as one of the most important constructional rudiments of a transformer, and any weakness of insulation would lead to a transformer failure. To check the insulation integrity, a sensitive technique like the standard impulse test is used. Testing of winding insulation is performed by comparing the voltage and current waveforms of reduced and full test voltages, previously. Transfer function has been exercised via applying fast Fourier transform to analyze the transformer state in the frequency domain. In order to enhance the fault detection sensitivity in transformer impulse test, this paper proposes a new method using short time Fourier transform in the transfer function analysis, derived from experimental works on 25 kVA and 26 MVA transformers. It is concluded that the fault detection sensitivity is improved significantly by using STFT analysis, including minor faults.