A New Fault Location Method

by Murari Saha, ABB, Sweden , Eugeniusz Rosolowski and Jan Izykowski, Wroclaw University of Technology, Poland

## Introduction  Algorithms for accurate location of faults on power lines have been a subject of great interest of researchers since the power system reliability became an important factor for network operators and customers. Accurate location is especially of utmost importance for series-compensated lines, which are spreading over few hundreds of kilometers and are vital links between the energy production and consumption centers. Among the known methods, the approach based on an impedance principle is the most popular. In particular, the algorithms utilizing one-end current and voltage measurements have been presented. In turn, in use of two-end currents and voltages, measured synchronously with the aid of PMUs, has been considered. The use of the un-synchronized measurements has been proposed in. The method for application with current differential protective relays of a single series-compensated line was introduced in. The other fault location techniques for such lines are based on knowledge based approaches. In this article a new impedance -based fault location algorithm for series-compensated lines (Figures 2a and 2b) is presented. It is considered that a compensation is accomplished with fixed series capacitors (SCs) - Figures 2a and 2b. The SCs are equipped with MOVs (Metal Oxide Varistors) for overvoltage protection. The other details for the capacitor bank, as not important for the conducted considerations, are not shown here. The approach for a single series-compensated line was taken for further development and adaptation to get a novel fault location algorithm.  The algorithm is formulated for a more general case of a double-circuit series compensated line (Figure 2b).

Innovative contribution relies on using the following specific in-complete two-end measurements: - phasors {IAA}, {IBA} of three-phase currents from both ends of the faulted line circuit measured by protective differential relays, - three-phase voltage {vA} measured locally, i.e.: at the bus AA, where the fault locator FLA is consider as to be embedded into the relay DIFF_RELA, - zero-sequence current iAB0 from the healthy parallel line (provided for compensating for the mutual coupling between the line circuits). The fault location function can be embedded into the differential relay at one end only, or into the relays at both line ends (Figures 2a and 2b). In the latter case much superior fault location is achieved and therefore it is taken for further considerations. The algorithm is derived for a double-circuit line (Figure 2b), however, after cancelling the mutual coupling elements it suits to a single line as well.
A fault is of a random nature and can appear at any line section, i.e. between the bus AA and the capacitor bank (fault FA) or between the bus BA and the bank (fault FB), as shown in Figures 2a and 2b.

Therefore, two subroutines: SUB_A (Section 2), SUB_B (Section 3) are utilized for locating these hypothetical faults FA and FB, respectively. The final result is selected with use of the selection unit (Section 4).

Section 2: Fault Location Subroutine SUB_A  - In relation to Figure 3 the following generalized fault loop model for fault FA can be stated: (1)

where:  dFA - unknown distance to fault [p.u.]; RFA - unknown fault resistance; VAp, IAp - fault loop voltage and current; IFA - total fault current (fault path current); Z1LA - positive-sequence impedance of the line section AA-X;

Note: , where: Z1L - positive-sequence impedance of the whole line AA-BA, dSC - relative distance from bus AA to SCs&MOVs.

Fault loop voltage and current are composed accordingly to the fault type as follows: (2) (3)

where:   - weighting coefficients (given in Table 1); 1, 2, 0 - digits in subscripts used for denoting positive-, negative- and zero-sequence components of the signals, - zero-sequence impedance of the line section AA-X, - zero-sequence mutual coupling impedance for the section AA-X.

In order to minimize an influence of line shunt capacitances on accuracy of determining the total fault current involved in (1), it is not calculated by direct adding the phase currents at both ends of the faulted line.  Instead, it is calculated using the following generalized fault model: (4)

where:   - share coefficients (Table 2);   - sequence components of currents at the bus AA;   - sequence components of currents at the bus BA.

Note that in (4) instead of using the fault quantities:  , the superimposed currents are applied: , (5)

where the subtracted currents are taken from the pre-fault time interval (superscript: pre).

Usage of the superimposed quantities (5), and not the fault quantities, is advantageous since lower errors arise due to neglecting line shunt capacitances. The recommended share coefficients (Table 2) assures that the zero-sequence currents are not involved ( ) in total fault current calculations.

After resolving (3) into the real and imaginary parts, and eliminating the unknown fault resistance (RFA), the sought fault distance (dFA) is determined as: (6)

Having the fault distance calculated (6), the fault resistance RFA can be also determined, for example as: (7)      