This paper develops a novel graph convolutional network (GCN) framework for fault location in power distribution networks. The proposed approach integrates multiple measurements at different buses while taking system topology into account. The effectiveness of the GCN model is corroborated by the IEEE 123 bus benchmark system. Simulation results show that the GCN model significantly outperforms other widely-used machine learning schemes with very high fault location accuracy. In addition, the proposed approach is robust to measurement noise and data loss errors. Data visualization results of two competing neural networks are presented to explore the mechanism of GCN's superior performance. A data augmentation procedure is proposed to increase the robustness of the model under various levels of noise and data loss errors. Further experiments show that the model can adapt to topology changes of distribution networks and perform well with a limited number of measured buses.

Fault Location in Power Distribution Systems via Deep Graph Convolutional Networks

Transmission lines are one of the most widely distributed engineering systems meant for transmitting bulk amount of power from one corner of a country to the farthest most in the other directions. The expansion of the lines over different terrains and geographic locations makes these most vulnerable to different kinds of atmospheric calamities which more often develops faults in line. It is imperative to remove the faulty line at the earliest to restrict undue outflow of bulk power through the faulted point as well as restore system stability earliest to resume normal power flow operation. Here lays the importance of having a robust fault identification, classification and localization algorithm which would be successfully able to drive as well as actuate the digital relaying system. Researchers have worked out several methodologies in developing improved power system protection algorithms which would be able to serve to eliminate faults immediately on occurrence of the same. A brief yet exhaustive review has been presented in this article including the several methodologies adopted by numerous researchers for developing effective fault diagnosis schemes, mentioning about the highlights as well as the shortcoming of each of the methods. This compact and effective survey of literature works would help researchers to take up appropriate techniques for different purposes of transmission line fault analysis.

Transmission Line Faults in Power System and the Different Algorithms for Identification, Classification and Localization: A Brief Review of Methods

This paper presents a new fault location algorithm for radial distribution networks employing synchronized distributed voltage traveling wave (TW) observers. A robust and accurate fault location algorithm significantly improves the distribution networks reliability and reduces the risk of bush fires and electrocution resulting from sustained undetected faults. The medium voltage distribution networks include numerous junctions and many shunt and series connected devices, such as capacitor banks, transformers and cables, which makes fault location far more complicated. This paper investigates the effect of power system components on the propagation of traveling waves and proposes a method for a fault location in heavily branched radial distribution feeders. Results show that parasitic shunt capacitances in transformers have a significant impact on traveling time of incident waves to the location of the TW observers and compensation for this effect will improve the accuracy of fault location.

Fault Location on Radial Distribution Networks via Distributed Synchronized Traveling Wave Detectors

This paper develops a novel graph convolutional network (GCN) framework for fault location in power distribution networks. The proposed approach integrates multiple measurements at different buses while taking system topology into account. The effectiveness of the GCN model is corroborated by the IEEE 123 bus benchmark system. Simulation results show that the GCN model significantly outperforms other widely-used machine learning schemes with very high fault location accuracy. In addition, the proposed approach is robust to measurement noise and data loss errors. Data visualization results of two competing neural networks are presented to explore the mechanism of GCNs superior performance. A data augmentation procedure is proposed to increase the robustness of the model under various levels of noise and data loss errors. Further experiments show that the model can adapt to topology changes of distribution networks and perform well with a limited number of measured buses.

Single-phase ground fault location method for distribution network based on traveling wave time-frequency characteristics

Fault location in distribution networks is difficult for multiple discontinuities, such as branches and junction points. This paper proposes a fault location scheme using network topology information and reclosure-generating traveling waves. Based on the topology, the circuit breaker closure-generating normal traveling waves (CNTWs) can be analyzed. When permanent faults occur, the circuit breaker reclosure-generating fault traveling waves (RFTWs) contain the information on fault position. The difference between the CNTWs and RFTWs, defined as reclosing superimposed traveling waves (RSTWs), is calculated. For RSTWs, their initial traveling waves are detected by wavelet transform. The time difference between the reclosing instant and the arrival instant of the reflected traveling wave of the fault point is employed to calculate the fault distance. Moreover, due to large numbers of branches, a fault distance may correspond to many sections. To determine the fault section, a database which contains the RSTWs induced by the faults occur in each section is built. The fault section can be identified by finding the most similar signals in the database by taking advantage of an improved dynamic time warping method. The simulation results indicate that the scheme can locate fault for distributed feeders regardless of grounding system type, load variation, and fault type.

Fault Location for Radial Distribution Network via Topology and Reclosure-Generating Traveling Waves

Fault location in distribution systems is difficult for multiple discontinuities, such as branch and junction points on feeders. This paper proposes a scheme for fault location in distribution systems using network topology information and circuit breaker reclosure-generating travelling waves. Based on the topology information, the circuit breaker closure-generating travelling waves can be got through analysis. When a permanent fault occurs, the circuit breaker reclosure-generating travelling waves contain the information on fault position. In this paper, the difference between the pre-fault travelling waves and post-fault travelling waves induced by the breaker reclosing in a feeder, defined as reclosing superimposed travelling waves, is calculated. For the reclosing superimposed travelling waves, the initial travelling wave reflects the fault distance. Subsequently, wavelet transform is applied to extract the travelling wave reflected from the fault point. Finally, the time difference between the reclosing instant and the arrival instant of the reflected travelling wave is employed to calculate the fault distance. In order to verify the effectiveness of the proposed scheme, several simulations were carried out using ATP-EMTP software. The simulation results indicated that the proposed scheme can be effective in fault location for distributed feeders, regardless of grounding system type.

Travelling waves-based fault location scheme for feeders in power distribution network

An increasing number of extra-high-voltage (EHV) and ultra-high-voltage (UHV) transmission lines have motivated the need for ultra-high-speed line protection. Traveling-wave directional protection has the advantages of ultra-high-speed operation and high sensitivity, but its reliability is doubted due to its dependence on detecting the wave-heads of traveling waves and the limitation of the limited bandwidth of the coupling capacitive voltage transformer. Incremental-quantity directional protection is another ultra-high-speed alternative. Aiming at improving the reliability of a directional relay (namely the core of a directional protection) without compromising its ultra-high-speed operation, this paper proposes a novel ultra-high-speed directional relay based on correlation of incremental quantities. The incremental voltage and current within a specific window only consist of some specific fault-generated traveling waves. Therefore, the incremental voltage is in approximately direct proportion to the incremental current. The proportional coefficient is positive or negative in the case of a reverse fault or a forward fault, respectively. The correlation coefficient of the incremental voltage and current, processed by an averaging filter, is used to identify the fault direction. Extensive simulations verified that the proposed directional relay can serve as an ultra-high-speed and reliable directional relay for non-, shunt-, or series-compensated EHV/UHV transmission lines.

An Ultra-High-Speed Directional Relay Based on Correlation of Incremental Quantities

Equivalent traveling waves based current differential protection of EHV/UHV transmission lines

Current travelling wave based single-ended fault location method needs to recognize the travelling waves propagating from the fault direction as well as the travelling wave reflected from the fault point or remote-end bus. This paper firstly compares the current travelling waves of the fault line and the longest health line connected to the local bus based on dyadic wavelet transform to obtain the current travelling waves propagating from the fault direction. Secondly, based on the cross transmission phenomenon of modal travelling waves under the scenario of single-phase-to-ground fault, the arrival sequence relation and the polarity relation of aerial-modal component and zero-to-aerial-modal component (i.e. cross transmission component) of the current travelling waves reflected from the fault point and the remote-end bus are investigated and the source of the second current travelling wave propagating from the fault direction is identified. Thirdly, the first and the second current travelling waves propagating from the fault direction are utilised to locate the single-phase-to-ground fault. Finally, a large number of EMTP based simulation cases are used to test and validate the proposed fault location method.

Aoyu Lei

Papers

Fault Location for Radial Distribution Network via Topology and Reclosure-Generating Traveling Waves

Travelling waves-based fault location scheme for feeders in power distribution network

An Ultra-High-Speed Directional Relay Based on Correlation of Incremental Quantities

Equivalent traveling waves based current differential protection of EHV/UHV transmission lines

A novel current travelling wave based single-ended fault location method for locating single-phase-to-ground fault of transmission line