Different automatic fault location approaches have evolved in order to make the grid smarter. Therefore, there is a need to optimally select the fault location approach depending on the type of data available.

Fault location in power systems is part of the restoration process following the occurrence of a fault. Shorter restoration times will be achieved if the fault is located promptly thus improving the reliability indices of the utilities such as System Average Interruption Duration Index (SAIDI) and System Average Interruption Frequency Index (SAIFI) to enable them to meet regulatory requirements. This is one of the advantages of smart grids over traditional grids, in addition to automation capability. Automation can easily be justified in view of regulators requiring reliability improvement, given the monetary penalties and rewards associated with poor and good reliability respectively.

Traditional fault location

Traditionally, SCADA (supervisory control and data acquisition) systems in substations use remote terminal units (RTUs) to acquire analog and digital measurements of branch flows, bus voltages, breaker status, frequency, and transformer tap positions, etc. Such measurements are sent to the energy management system (EMS) every two to ten seconds. Network operators use such information to learn when faults occur [1].

A large section or an entire feeder may be affected during a fault, and the exact location of the fault is not readily identified by the network operators. The maintenance crew is then dispatched to find and fix the fault. This requires trial and error switching actions of circuit breakers and isolators to pin-point the exact area of the fault. This can take a considerable amount of time during the day but even longer at night or in poor weather, and increases the duration of time that customers are left without supply [2].

Automated fault location

With advancement in technology, intelligent electronic devices (IEDs), other than the RTU, have been developed. These have a higher sampling rate, better accuracy and can store more data. Proper utilisation of such acquired data can enhance the prediction, monitoring and post mortem analysis of power system events. Fault location methods can also benefit from the large amount of data made possible by these IEDs in smart grids [1]. Figure 1 shows a comparison of the outage restoration process without and with automation.

Smart grids use sensors and communication tools through fault location schemes to transmit information on the location of a fault when one occurs and thus reduce the time that is spent trying to find the exact location of the fault and therefore reduce the service interruption duration.

Fault location methods

Distance relays can be used to locate faulted areas reliably but more accurate methods are required for fault location.

Transmission lines method
Calculation of the location of faults on transmission lines can be done using voltage and current power frequency components or higher frequency transients generated by the faults. Phasor-based methods which use the fundamental component of the signal and the lumped parameter model of the line or time-domain based methods which use the transient components of the signal and distributed parameter model of the line can be used. Depending on the availability of recorded data, we can divide each of these two categories into two broader categories of single-end methods, and double-end methods.

In single-end methods, the data from one terminal only of the line is available for use while in doubleend methods, data from both or multiple terminals are available for use. In double-end methods, synchronised or unsynchronised phasor measurements and samples can be used; they are more accurate than the single-end method but require more communication for data synchronisation. The location of the fault can be through the calculation of the impedance of the line or be based on travelling waves, which use transient signals generated by the fault [1].

Single-ended impedance based methods are simple and fast but the simplest approach is the reactancebased method in which the resistance of the line is ignored and may result in errors due to remote end infeed currents, load impedance, power transmission angle and angle difference between line and source impedances. In double-end methods, the voltage measurements of the fault point from the two ends of the line are equalised. For either phasor or time based impedance focused methods, the distance to fault is estimated as a function of the total line impedance assuming a homogenous line [1].

Generally, digital fault recorders (DFRs) and other IEDs are placed in critical substations; therefore data may not be available for some lines even though they have protective relays. Such cases may require the use of some unconventional methods based on wide area measurement systems (WAMS).

Optimised transmission fault location method

The algorithm in Figure 2 can be used to select an optimal fault location scheme for a transmission line based on the type of available data and whether the fault type is known.

Distribution feeders method

Fault location in distribution networks is generally more complex than in transmission networks due to different types of conductors and structures, lateral branches, load distributed along feeders and frequent network reconfiguration [4]. Therefore, methods proposed for transmission may not be easily applicable in distribution networks. Different fault location methods based on the type of data used in locating the fault include:
i. Apparent impedance measurement;
ii. Direct three phase circuit analysis;
iii. Superimposed components;
iv. Travelling waves;
v. Power quality monitoring data; and
vi. Artificial intelligence.

The merits and demerits of these methods can be found in IEEE Transactions on Smart Grid, Vol. 2, No. 1, March 2011. One of the fault location methods includes the placement of fault passage indicators along feeders with the possibility of remote data collection at main substations or distribution control centres [4]. In this system, sensors placed along the feeder length are used to determine whether a fault current has passed through a feeder section. It is more accurate than a distance to fault estimator, which uses volt drop or impedance based fault location methods to estimate the distance between the substation and a fault, which does not work well in radial distribution feeders with many lateral sub-feeders. Wireless or wired communication methods can be used in both of these systems to send data to the repair crews or the network operator.

On the other hand, there are main substation located schemes based on algorithms that use voltage and current signal measurements obtained from IEDs. These include digital transient recorders or digital protection equipment, PMUs and smart meters. Such schemes require additional information including the characteristics of protection devices and their locations, load profiles and the configuration of the network, which can be obtained from databases in the main substation or distribution control centres [4].


The use of fault location methods with IEDs provide better monitoring than SCADA. An application of such systems in transmission and distribution networks will require the installation of sensors and IEDs along the feeders and the upgrading of the communications systems, which involve much infrastructure investment and money. However the improvement in reliability indices and savings in customer damage costs due to the reduction in outage times, especially for feeders and load points with industrial or commercial customers, may justify the investment. ESI

About the authors

Kehinde O Awodele is a registered electrical engineer (COREN, MNSE, MIEEE) with an MSc (Eng) Electrical Power and Machines. Currently she is a lecturer in the Department of Electrical Engineering at the University of Cape Town, South Africa. She worked in the electricity meter manufacturing industry for several years in several capacities. Her research interests include Power system reliability, Smart grids, Distributed generation, Renewable energy and Demand response.

Elvis Leke Alembong has held a B.Engr. degree in Electrical and Electronic Engineering (Power System Engineering) from the University of Buea, Cameroon since 2015 and is currently an M.Sc Engineering research candidate in the area of Substation automation and protection improvement using IEC 61850 in the Department of Electrical Engineering, University of Cape Town.

[1] M. Kezunovic 2011. Smart Fault Location for Smart Grids. IEEE Transactions on Smart Grid.
2(1):11-22. DOI:10.1109/TSG.2011.2118774.
[2] P. P. Subban & K. O. Awodele 2013. Reliability Impact of Different Smart Grid Techniques on a
Power Distribution System. 2013 IEEE PES Conference on Innovative Smart Grid Technologies
ISGT Latin America). 1.10.1109/ISGT-LA.2013.6554467.
[3] D. M. Staszesky, D. Craig and C. Befus, “Advanced Feeder Automation is Here”, IEEE Power and
Energy Magazine, September/October 2005, pp. 56-63.
[4] S. Kazemi, “Reliability Evaluation of Smart Distribution Grids”, Doctoral Dissertation,
Department of Electrical Engineering, Aalto University, Espoo, 2011.