DocumentCode
2608613
Title
Dielectric loss angel measurement algorithms based on DFT spectrum correction
Author
Jing, Lei
Author_Institution
Shapingba Electr. Power Bur., Chongqing
fYear
2008
fDate
9-12 Nov. 2008
Firstpage
433
Lastpage
436
Abstract
It is well know that, in discrete Fourier transform(DFT) based waveform analysis of multifrequency signals, there will be great dielectric loss measurement error. In this paper, the spectrum leakage presented in details, the energy of the signals will transpire because of the frequency fluctuation which brings asynchronous sampling, and the energy correlation is not zero between different frequency points. On the basis, smoothing signals steep degrees by using cosine window, in order to improve phase correction on discrete spectrum, two different methods which adopt frequency offset to measure the signal phase on the maximum spectrum peak are presented. By MATLAB simulating, the results show that two methods are easily carried out with high accuracy and Practicality.
Keywords
condition monitoring; correlation methods; dielectric loss measurement; discrete Fourier transforms; insulation testing; signal sampling; smoothing methods; spectral analysis; DFT spectrum correction; MATLAB simulation; asynchronous sampling; cosine window; dielectric loss angel measurement algorithm; discrete Fourier transform; energy correlation; frequency fluctuation; insulation condition monitoring; measurement error; multifrequency signal; signal phase measurement; signal smoothing; spectrum leakage; waveform analysis; Dielectric loss measurement; Dielectric losses; Dielectric measurements; Discrete Fourier transforms; Fluctuations; Frequency; Loss measurement; Measurement errors; Sampling methods; Signal analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
High Voltage Engineering and Application, 2008. ICHVE 2008. International Conference on
Conference_Location
Chongqing
Print_ISBN
978-1-4244-3823-5
Electronic_ISBN
978-1-4244-2810-6
Type
conf
DOI
10.1109/ICHVE.2008.4773966
Filename
4773966
Link To Document