Classification of faults and switching events by inductive reasoning and expert system methodology

Abstract

Under certain conditions, one electrical parameter (independent variable) is not enough to detect high-impedance faults on certain surface conditions. These faults do not draw sufficient current for detection and may draw less current than similar faults on other soil surfaces. Moreover, because every electrical detection parameter displays characteristics of randomness, it is difficult to assign a probability that a given event is a high-impedance fault, rather than a switching event. It has been shown that detection by induction laws can improve the classification of faults and switching events. The second and third laws of induction are utilized with a minimum entropy method. Setting detection threshold values using induction methods is also proposed. The methods presented are taken from ongoing research in high-impedance fault detection. While the techniques have not been reduced to practice or field-tested, they hold promise for future improvements in the relaying of high-impedance faults.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>