Abstract--Modernizing the power distribution system implies improving the reliability and performance of protection devices. By incorporating data-mining in the process of designing protection functions, the limits of performance are extended. We propose a method that uses data-mining, able to detect high impedance faults (HIFs) in multi-grounded distribution networks when conventional devices are insufficient. HIFs are produced when overhead lines contact a quasi-isolated surface, such as a tree or the ground. The fault current can be lower than the residual current under normal conditions; hence overcurrent devices do not detect this fault. We describe a set of indicators that characterize HIFs and that can be used in data-mining to distinguish fault situations from other situations. The result is a HIF detection function whose development is based on pattern recognition analysis. The presented methodology can be applied to other fault detection problems to achieve more reliable protection devices.

No abstract

Abstract--Loop operation of distribution networks is more and more common. The increasing penetration of distributed generation and higher power quality requirement are making the loop structure more interesting. We study the impact of the unbalanced phase impedances. Measurements and simulations show that it creates a mutual coupling between the positive sequence and the zero sequence system. Balanced load creates a zero sequence voltage between two feeders. Their connection of the both ends to form a loop results in circulation of a zero sequence current. In case of an earth fault this current will be added to the fault current and can negatively influence the protection devices. Healthy lines do not have active current in the zero sequence system because there is only capacitance connection with the earth but the circulating current change this hypothesis. We propose a study of this problematic current for simple and multi loop in distribution networks.

Single phase earth fault in compensated network creates faulty current with very small magnitude. The connection of a Petersen Coil in the transformer neutral turns the zero sequence impedance into very high ohmic system. The small faulty current makes the detection of the faulty feeder impossible with simple over current protection. That is why algorithms are using transients. This paper presents a novel method to detect such earth fault with a strong sensitivity regarding the fault impedance. The algorithm builds a model of the sound feeder behaviour using the zero sequence voltage and current. The healthy feeder can be considered as a zero sequence capacitance. By measuring the deviation of a perfectly capacitive feeder, the faulty one can be detected. Algorithm description and tests are presented with comparison of small impedance fault and high impedance with recordings and simulations.

No abstract