Transient Earth Fault Detection - GE MiCOM P40 Agile Technical Manual

P543i/P545i
9

TRANSIENT EARTH FAULT DETECTION

Some distribution systems run completely insulated from earth. Such systems are called unearthed systems. The
advantage of an unearthed system is that a single phase to earth fault does not cause an earth fault current to
flow. This means the whole system remains operational and the supply is not interrupted. The system must be
designed to withstand high transient and steady state overvoltages, however, and so its use is generally restricted
to low and medium voltage distribution systems.
When there is an earth fault in an unearthed 3-phase system, the voltage of the faulted phase is reduced to the
earth potential. This causes the phase voltage in the other two phases to increase, which causes a significant
charging current between the phase-to-earth capacitances. This can cause arcing at the fault location. Many
systems use a Petersen coil to compensate for this, thus eliminating the arcing problem. Such systems are called
compensated networks. The network is earthed with an inductive reactor, where its reactance is made nominally
equal to the total system capacitance to earth. Under this condition, a single-phase earth fault does not result in
any steady state earth fault current.
The introduction of a Petersen coil introduces major difficulties when it comes to determining the direction of the
fault. This is because the faulted line current is the sum of the inductive current introduced by the Petersen coil
and the capacitive current of the line, which are in anti-phase with each other. If they are equal in magnitude, the
current in the faulted line is zero. If the inductive current is larger than capacitance current, the direction of the
faulted line current will appear to be in the same direction as that of the healthy line.
Standard directionalizing techniques used by conventional feeder protection devices are not adequate for this
scenario, therefore we need a different method for determining the direction of the fault. Two commonly used
methods are the First Half Wave method and the Residual Active Power method.
First Half Wave Method
The initial transient wave, generated at the fault point travels towards the bus along the faulted line, until it
reaches the healthy line. For forward faults the high frequency fault voltage and current components are in
opposite directions during the first half wave, whereas for reverse faults, they are in phase. This fact can be used to
determine the fault direction. This method, however, is subject to the following disadvantages:
The time duration of the characteristic is very short, in most cases not more than 3 ms. Because of this, it
●
requires a high sampling frequency (3000Hz or even higher)
It requires an analogue high pass filter, necessitating special hardware
●
It is affected by the fault inception angle. For example, when the fault inception angle is 0°, there are no
●
initial travelling waves.
Residual Active Power Method
Residual Active power, which is sometimes used to detect the instance of a fault can also in some cases be used
for detecting the fault direction. Although the capacitive currents can be compensated by an inductive current
generated by a Petersen coil, the active (instantaneous) current can never be compensated for and this is still
opposite to that of the healthy line. This fact can also be used to directionalise the fault.
For a forward directional fault, the zero-sequence active power is the power loss of Petersen's coil, which is
negative. For a reverse fault, the zero-sequence active power is the power loss of the transmission line, which is
positive. This method, however, is subject to the following disadvantages:
The zero-sequence active power will be very small in magnitude for a reverse directional fault. Its value
●
depends on the power loss of transmission line.
The zero-sequence active power may be too small in magnitude to be detected for a forward directional
●
fault. Its value depends on the power loss of Petersen coil.
High resolution CTs are required
●
Due to the low magnitude of measured values, reliability is compromised
P54x1i-TM-EN-1
Chapter 13 - Current Protection Functions
391