Determining the locations of faults in distribution systems
Date
1998-01-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
Type
Degree Level
Doctoral
Abstract
The conventional approach for estimating the locations of transmission line faults has been to measure the apparent impedance from a line terminal to the fault and to convert the reactive component of the impedance to line length. Several methods, that use voltages and currents measured at one or both line terminals, have been proposed in the past. Methods for locating faults on radial transmission lines and rural distribution feeders have also been suggested. These methods do not adequately address the problems associated with fault location on distribution systems that have single or multiphase laterals and/or tapped loads. A technique that estimates the location of a shunt fault on a radial distribution system that has several single and/or multiphase laterals has been developed. Load taps and non-homogeneity of the system are taken into account. The apparent location of a fault is first estimated by computing the impedance from the fundamental frequency voltage and current phasors, and converting the reactive component of the impedance to line length. The sequence voltages and currents at the fault are expressed as functions of the distance to the fault as well as the impedances of loads beyond the fault. The expression for the imaginary component of the fault impedance is equated to zero and the resulting equation is solved using an iterative approach. Multiple estimates may be obtained for a fault in a distribution system that has laterals. One of the estimates is identified as the most likely fault location by using information from fault indicators which are strategically placed on the laterals. The developed technique, which can handle single-phase-to-ground, two-phase-to-ground, phase-to-phase and balanced three-phase faults was tested to evaluate its suitability. Results from computer simulations of faults indicate that the proposed technique is more accurate than the reactive component method. Studies also demonstrate that the sensitivity of the proposed technique is comparable to that of the reactive component method. A prototype fault location system was also developed. The system was tested using simulated voltage and current waveforms. Results show a close agreement with those obtained from the non-real time tests.
Description
Keywords
Citation
Degree
Doctor of Philosophy (Ph.D.)
Department
Electrical Engineering
Program
Electrical Engineering