Earth-Fault Location - Allen-Bradley Rockwell Automation 857 User Manual

Earth-Fault Location

The 857 includes a sophisticated stand-alone earth-fault location
algorithm. The algorithm can locate an earth-fault accurately in
radically operated compensated earthed networks.
The function can locate a fault only if the fault resistance is low, say
less than 50 ohms. The fault location is given in reactance value.
This value can then be exported, for example, with event to a DMS
(Distribution Management System). The system can then localize the
fault and display it on a map.
The fault location must be used in the incoming bay of the
substation. Therefore, the fault location is obtained for the whole
network with just one 857. This is very cost-effective upgrade of an
existing system.
Please note also that the earth-fault location function requires a
change during an earth-fault. This change is done by switching the
secondary resistor of the compensation coil on or off. The fault
should be allowed to be on at least 200 ms, of which 100 ms without
the resistor. The resistor change can be done by using the logic
functionality of the 857.
The reactance value is converted to distance in the DMS. The
following formula is used:
3 * X
=
s
+ +
Xo X X
1 2
s = distance in km
X = reactance calculated by the 857
X = zero sequence reactance per kilometer of the line
0
X = positive sequence reactance per kilometer of the line
1
X = negative sequence reactance per kilometer of the line
2
The algorithm functions in the following order:
1. The needed measurements (phase currents and voltages) are
continuously available.
2. The fault distance calculation can be triggered in two ways: by
switching ON or OFF the secondary resistor (that is, by using a
digital input) or the calculation can be triggered if there is a
change in earth fault or negative sequence current
3. The fault phase is identified by that the voltage of the faulted
phase is decreased at least by half.
4. The fault distance is calculated by dividing the change of the
voltage by the change of the negative sequence current.
5. Only the imaginary part is used, so then the reactance is solved.
Supporting Functions
Where,
4-29
857-UM001A-EN-P – July 2009