Distinguishing between Perfect and Faulty Insulators by Bispectrum and Classifier
Wen Kang-zhen, Yuanfang Wen, Jian-Shuang Lu · 2010
According to analyzing the datum of the leakage currents from both perfect and faulty insulators that are at the state of various dirtiness, humidity, and temperature, and are applied non-sinusoidal voltages at different virtual values, three results are obtained. First, the bispectrum of the leakage current from contaminated insulator is mostly related to insulator; not to dirtiness. Bispectrum is the eigenvalue sensitive to distinguishing between perfect and faulty insulators. Secondly, a three-dimensional vector, which consists of the v of bispectrum, Shannon entropy, and wave crest factor, can divide all the insulators into two not only non-overlapping but far apart regions. One is the region of perfect insulators. The other is the region of faulty insulators. Thirdly, using above a three-dimensional vector as input vector, any of three classifiers, Euclide distance judgment, grey relational analysis of B-mode, and BP artificial neural network, can exactly distinguish between perfect and faulty insulators at various dirtiness, humidity, temperature and the virtual values of the voltage involving harmonics.