Abstract
This paper presents a method for locating ground faults in ungrounded and high resistance grounded networks. In such networks, the contribution of the fault to the line currents is minimal, thus creating a challenge to the popular impedance-based fault locators employed in the industry. A compensation method is introduced to alleviate this impediment to the use of impedance-based fault locators; the compensated signals are consistent with the transmission line wave equations and are formed by the fault signal components thus making the fault information more visible. The proposed approach employs (i) phasor extraction based on forward and backward finite impulse response (FIR) filtering to better capture the fault signal dynamics, (ii) an initialization stage employing redundant measurement equations of the fault voltage, and (iii) an augmented matrix approach for least-squares estimation that models the physical conditions of the fault in symmetrical component coordinates. The proposed method is contrasted with a recent transient analysis technique for locating ground faults in non-solidly grounded systems and a synchronized measurement technology-based fault locator; the numerical results show a significant improvement in accuracy while employing standard data acquisition devices.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have