CN112119556B

CN112119556B - Method and apparatus for use in ground fault protection

Info

Publication number: CN112119556B
Application number: CN201980032077.8A
Authority: CN
Inventors: 阿里·瓦尔罗斯; 扬内·阿尔托宁
Original assignee: ABB Schweiz AG
Current assignee: ABB Schweiz AG
Priority date: 2018-05-18
Filing date: 2019-05-17
Publication date: 2023-01-24
Anticipated expiration: 2039-05-17
Also published as: CN112119556A

Abstract

A method and apparatus for use in ground fault protection in a three-phase electrical network, the apparatus being configured to: detecting a relative earth fault (60) in the network (50); for each of the phases (a, B, C) of the network, determining the phase current during the fault or the change in the phase current due to the fault; detecting a faulted phase of the network; determining an estimate of the ground fault current based on the fault phase and the phase current or a change in the phase current; determining zero sequence voltage or change of the zero sequence voltage of the power grid; and determining the direction of the relative earth fault from the measurement point (40) based on the estimate of the earth fault current and the zero sequence voltage or the change in the zero sequence voltage.

Description

Method and apparatus for use in ground fault protection

Technical Field

The present invention relates to a method and an apparatus for use in earth fault protection in a three-phase power network.

Background

Ground fault protection functions in high impedance grounded networks, such as compensation networks, ungrounded networks or high resistance grounded networks, may be based on zero sequence voltages to the network

(or neutral point voltage, residual voltage)

)

And the residual current (or sum current) at the measuring point

The measurement of (2). Monitoring neutral point voltage

And comparing it with a predetermined set value can be used as a general indication of a ground fault somewhere in the current-connection network. However, the detection of faulty and normal feeders or fault directions as generally seen from the measurement point cannot be based solely on the neutral voltage

But need to be additionally, for example, for residual currents

Is measured. Furthermore, especially in compensation networks and ungrounded networks, the magnitude of the residual current may not be a selective indication of the fault direction and more advanced methods may be required.

There may be selectivity requirements for the ground protection function. This means that for example only "true" faulty feeders or faulty line segments of feeders should be detected as faults. For example, a normal feeder or a normal line segment of a feeder should not be detected as a fault to avoid unnecessary disconnection of the normal part of the network, resulting in undesired interruptions for e.g. the end customer.

Some of the actually used protection functions applied in high impedance grounding networks can be divided into four groups:

1. current-based method

2. Power-based method

3. Admittance based method

4.1 to 3 in any combination

In the current-based approach, the operating quantity of the protection is the residual current

The phasor, its magnitude, real or imaginary part or phase angle may be compared to a predetermined threshold. In a power-based approach, the amount of operation of the protection is the residual power

Phasors

Its magnitude, real or imaginary part or phase angle may be compared to a predetermined threshold. And in the admittance-based approach, the operative amount of protection is the residual (neutral) admittance

Phasors

Its magnitude, real or imaginary part or phase angle may be compared to a predetermined threshold. Especially in compensation networks and ungrounded networks, the amplitude criteria are usually not selective indications of the direction of the fault.

In compensation networks, e.g. residual currents measured at the beginning of the feeder

May not normally equal the ground fault current flowing from the faulted phase to ground at the fault location

The residual current is typically only a fraction of the ground fault current, and the relationship between the residual current and the ground fault current can be written as:

or formula 1a

Wherein the content of the first and second substances,

equal to the portion of the ground fault current generated by the faulty feeder itself. During a complete ground fault (i.e., fault resistance equal to zero ohms)

The value of (c) can be approximated by using equation 2 (ignoring the natural resistive losses of the feed line itself):

wherein the content of the first and second substances,

is the earth fault current of the uncompensated feeder

ω＝2·π·f _n It is the nominal angular frequency of the network,

f _n is the nominal frequency of the network (e.g. 50Hz or 60 Hz)

C _0Fd Is the phase-to-earth capacitance (per phase) of the total feed line

U _PE Is the operating phase to ground voltage magnitude.

From equations 1a, 1b and 2, it can be concluded that: for example, in modern networks, especially in rural networks, as the share of installations (share) of underground cables increases,

may increase (for cables, C compared to overhead lines) _0Fd Is usually significantly higher), and

and

the difference between may thus become larger. This means a residual current

The ground fault current that may become worse and flow at the fault location

Is much worse.

The problem with solutions utilizing residual currents in the ground fault protection function is that the ground fault current is due to changes in the grid, such as any topology changes of the grid

May have different magnitudes, for example, due to faults or disturbances as well as fault locations and possible successful recovery processes. In addition, due to internal faults, e.g. in compensation coil tuning systems, earth fault currents

May have a magnitude greater than expected. In this case, the bucking coil may be severely detuned until the fault is detected and repaired or replaced. During such conditions, for example, residual current based ground fault protection may not be operated accurately and fast enough, which may pose a high risk to personal safety and equipment failure, for example.

Disclosure of Invention

It is an object of the present invention to provide a method and an apparatus for implementing the method so as to solve or at least alleviate the above problems or to provide an alternative solution. The object of the invention is achieved by a method, a computer program product and a device which are characterized by what is stated in the independent claims. Preferred embodiments of the invention are disclosed in the dependent claims.

The invention is based on the following idea: determining an estimate of the earth fault current through the measurement point at the point of the phase-to-earth fault based on the faulty phase of the three-phase electrical network and the determined phase currents of the three phases or based on a change in the faulty phase of the three-phase electrical network and the determined phase currents of the three phases, such that the estimate is based on the negative sequence current component; and then determining a direction of the relative earth fault from the measurement point based on the determined estimate of the earth fault current and the zero sequence voltage or based on the determined estimate of the earth fault current and the variation of the zero sequence voltage.

An advantage of the method and device of the invention is that the direction of the ground fault can be determined accurately, which increases the dependability of the ground fault protection.

Drawings

The invention will be described in more detail hereinafter by means of preferred embodiments with reference to the accompanying drawings, in which:

fig. 1 shows an example of a power grid according to an embodiment; and

fig. 2 shows a flow diagram according to an embodiment.

Detailed Description

The applications of the various embodiments described herein are not limited to any particular system, but may be used in conjunction with ground fault protection of various three-phase electrical networks. As an example, embodiments may be used in ungrounded networks with ungrounded or isolated neutral points, i.e. embodiments do not have an intentional neutral point grounding, but are only grounded through the natural relative ground capacitance of the network. As another example, embodiments may be used in compensation networks with compensated neutral points, also referred to as resonant grounding networks, where compensation of fault currents is achieved by installing one or more (Petersen) coils into the neutral point of the system. As yet another example, various embodiments may be used in an impedance grounding network having a neutral point provided with a resistive and/or reactive ground, such as a high resistive ground and/or a high reactive ground. In such a network with a high-resistance grounded neutral, the value of the ground resistance may be chosen such that its value substantially corresponds to the final capacitive reactance of the relative ground admittance (capacitance) of the current-connecting network, e.g. such that the ground fault current is limited to approximately the same or slightly larger value than the uncompensated capacitive ground fault current of the network. The electrical network with which the various embodiments are implemented may be, for example, an electrical transmission or distribution network or a component thereof, and may include several wires or segments. For example, the grid may have a radial configuration provided from one of its points or a loop configuration comprising one or more loops and provided from two or more points. Furthermore, the use of various embodiments is not limited to systems employing 50Hz or 60Hz fundamental frequencies or any particular voltage level.

Fig. 1 is a simplified diagram illustrating a power grid 50 in which various embodiments may be applied. The figure only shows the components necessary for understanding the various embodiments. An exemplary three-phase network 50 having phases a, B and C may be a medium voltage (e.g., 20 kV) distribution network fed through a substation comprising a transformer 10 and a bus 20. The exemplary network also includes wire outlets, i.e., feeders, one of which 30 is shown separately and is relatively admitted in the figure

And (4) showing. In addition to the line 30, other possible feeders and other network parts are referred to in the figure as being relatively admittance by earth

The "background network" 31 is shown. The network shown also comprises a compensation coil 70, which compensation coil 70 may be connected to the neutral point of the network, for example via a transformer 10. Admittance of the coil is

And the resistance in parallel with the coil is R _par 。

Is the current flowing through the coil 70. The figure also shows the inclusion of at least one protective relay at the connection point of the electric line 30 to the substation 10, 20The connection facilities of the electrical unit 41 and the point at which the phase a phase to earth fault 60 occurs. The term "phase-to-ground fault" herein generally refers to a single phase-to-ground fault. The protective relay unit 41 may be configured to detect the ground fault 60 based on suitable measurements and thus operate the ground fault protection of the electrical line 30. The operation of the ground fault protection may include tripping and/or preventing (blocking) tripping of one or more switching devices in the three-phase electrical network. Additionally or alternatively, the operation of the ground fault protection may include, for example, disconnecting or limiting the ground fault current of the ground fault 60 on the detected three-phase electric line 30 and/or performing an alarm. Disconnecting the ground fault current of the detected ground fault 60 may be performed by: the wires 30 are disconnected from the feed points, such as the

substations

10, 20, by means of suitable switchgear, such as circuit breakers or other kinds of switchgear (switchgears), which may be included in the connection facility. Limiting the ground fault current of the detected ground fault 60 may be performed by: the ground fault current flowing from the feed point to the wire 30 is limited or reduced without complete disconnection by means of a suitable fault current controller device. This may be based on controlling the coil 70 during a ground fault, for example. Such a limitation of the ground fault current may also be performed, for example, as a preparatory procedure prior to breaking the ground fault current. It should be noted that any number of feeders or other network elements may be present in the network. There may also be several feeding substations. Furthermore, the invention may be used with e.g. a switchyard without transformer 10. In the exemplary system of fig. 1, the functionality of the invention may be located, for example, in the relay unit 41. It is also possible to perform some measurements, for example only at the location of the relay unit 41, and then transmit the results to another unit or units (not shown in the figure) at another location for further processing. In other words, the relay unit 41 may be just a measuring unit, while the inventive function or a part of the inventive function may be located in another unit or units, possibly located at other positions.

The current and voltage values that may be required in the following embodiments may be obtained by suitable measurement facilities including, for example, current and/or voltage transformers at measurement points such as may be located at the position of the relay unit 41. For example, the voltage amount and the current amount may also be measured at different positions. In most existing protection systems, such values are readily available and therefore implementation of various embodiments does not necessarily require any additional measurement facilities or devices. However, how possible current and voltage values are obtained depends on the particular grid 50. For example, the phase currents I can be applied to the three-phase electrical lines 30 of the three-phase network 50 _A 、I _B 、I _C And/or other amounts of current and/or voltage that may be required in various embodiments, or monitoring of at least some of the amounts may begin only after a ground fault is detected, depending on, for example, whether a pre-fault value of the amount in question is required.

Fig. 2 shows a flow diagram according to an embodiment, an example of which is described below.

According to an embodiment, a phase-to-earth fault is detected 100 in the three-phase grid 50. Then, for each of the three phases A, B, C of the three-phase electrical network, the phase currents during the detected phase-to-ground fault or the change in the phase currents due to the detected phase-to-ground fault are determined at the measuring points in the three-phase electrical network 50 and the faulty phase of the three-phase electrical network is detected. The determination of the faulted phase may be performed using any known suitable method. An estimate of the earth fault current through the measurement point 40 at the point of the detected phase-to-earth fault 60 is determined 110 based on the faulty phase of the three-phase grid and the determined phase currents of the three phases or based on a change of the faulty phase of the three-phase grid and the determined phase currents of the three phases, such that the estimate is based on the negative sequence current component. In addition, a change in the zero sequence voltage of the three-phase system during the detected phase-to-earth fault or due to the detected phase-to-earth fault is determined 110. The determination of the zero sequence voltage or the change of the zero sequence voltage can be carried out in the determination of the estimate of the ground fault currentBefore, during and/or after. The determination of the zero sequence voltage or the change of the zero sequence voltage may be performed using any known suitable method. The zero-sequence voltage can be obtained, for example, from the open-delta winding of a three-phase voltage transformer, or can be calculated from the phase-to-ground voltage:

(the notation n refers to any integer greater than 1, a multiple of the network fundamental frequency). Next, the direction of the relative ground fault from the measurement point 40 is determined 120 based on the determined estimate of the ground fault current and the determined zero sequence voltage or the determined change in the zero sequence voltage. The direction of the fault generally refers to the direction of the location in the grid where the fault is located as seen from an observation point, such as a measurement point. As an example, if the measurement point 40 is located at the beginning of the wire outlet 30, for example, the direction of the fault may be towards the line outlet 30, i.e. the fault is located within the line outlet 30 (in the example of fig. 1, the arrow in combination with the relay unit 41 shows the fault 60 direction as being towards the line outlet 30), or the direction of the fault may be away from the line outlet 30, i.e. the fault is located within the

substation

10, 20 or within the background network 31. In this case, the determined direction of the fault may thus be used to determine whether a single feeder or line outlet including its possible branches or line segments is faulty or normal. Furthermore, in the case of two or more parallel feeders or line outlets, each of which is provided with a measuring point at the beginning thereof, it is possible to determine which of the two or more parallel feeders or line outlets is faulty.

According to an embodiment, ground fault protection in a three-phase electrical network is operated 130 based on the determined direction of the relative ground fault from the measurement point 40.

According to an embodiment, the operation of ground fault protection comprises: tripping and/or preventing tripping of one or more switching devices in a three-phase power grid; and/or to disconnect or limit the ground fault current of a detected phase-to-ground fault in the three-phase network. The operation of the ground fault protection may also include other or alternative measurements or actions, for example, made in accordance with the electrical system in question.

The detection 100 of a relative earth fault may be performed using any known method suitable for the type of network in question, such as a compensation network, an ungrounded network or a high resistance grounded network.

According to an embodiment, the detection 100 of the phase-to-earth fault comprises checking the zero sequence voltage amplitude (fundamental frequency, indicated by the superscript index 1) or at least two instants t ₁ 、t ₂ (where t is ₁ >t ₂ ) And comparing it with a predetermined threshold (the change may be calculated from the zero sequence voltage phasor or amplitude):

or

(change in phasor magnitude) or

(variation of amplitude)

Wherein the indices t1 and t2 are associated with different moments (t 1> t 2)

Monitoring of zero sequence voltage amplitude (fundamental frequency component amplitude exceeding threshold or at least two moments t) ₁ 、t ₂ The change in the amplitude between exceeds a threshold value, where t ₁ >t ₂ ) A very safe indication of a single phase to earth fault somewhere in the current connection network is given. An advantageous feature of zero sequence voltage is that there is no zero sequence voltage during non-fault related phenomena in the network, such as switching transients or inrush current events, which may confuse some other ground fault detection criteria.

The set value U0_ start _ threshold should preferably be set to a value as low as possible in order to maximize the fault detection sensitivity (in terms of fault resistance). However, to avoid false fault detection during normal state of the network, the detection threshold should preferably be set to a value (with margin) that is higher than the zero sequence voltage during normal state due to network relative admittance imbalance.

According to an embodiment, in a compensated network, the U0_ start _ threshold is not predetermined, but can be determined in real time. This can be performed, for example, by the arc suppression coil regulator (i.e., the controller of the arc suppression coil in the compensation network) or another unit or system connected thereto. The maximum value Uomax _ normal _ state of the zero sequence voltage due to the admittance imbalance of the system during the normal state is determined. In case the maximum value of the zero sequence voltage during the normal state is determined, then the U0_ start _ threshold value may be automatically determined in real time as:

u0_ Start _ threshold ≧ Uomax _ Normal _ State q0, where q0>1 is a user-defined safety margin. This embodiment may allow for an increased sensitivity for ground fault detection.

According to an embodiment, the detection 100 of the relative earth fault additionally or alternatively comprises: for triple negative-sequence current amplitudes (fundamental frequency only, fundamental frequency plus harmonics only or harmonics only) or at least two times t ₁ 、t ₂ (where t is ₁ >t ₂ ) Monitoring of a change in the amplitude; and comparing it with a preset threshold:

(with a settable pick-up delay),

or

(with a settable pick-up delay),

or

(with a settable pick-up delay),

wherein the content of the first and second substances,

n and m refer to frequency components, which are multiples of the network fundamental frequency.

n =1,m =1, or

n =1 and m = any integer greater than 1, or

n = any integer greater than 1, and m is any integer greater than n,

wherein the indices t1 and t2 are associated with different moments (t 1> t 2)

And is wherein the content of the first and second substances,

is based on the theory of symmetrical components at frequency f _n * n (n = any integer greater than or equal to 1) calculated negative-sequence current component, wherein,

since negative sequence currents may also be generated during non-fault related phenomena in the network, such as load unbalance (i.e. negative sequence current is a load-related and thus time-related quantity), switching surge conditions and saturation of the phase current transformers, the pick-up of the ground fault detection should preferably be set to a value higher than three times the magnitude of the negative sequence current measured during normal state of the network. Due to the fact that the normal state level may vary due to time-dependent properties of the load and topology changes of the network, the detection method may not be as sensitive as the fault detection method based on zero sequence voltages. The fault detection sensitivity based on the negative sequence current amplitude can be realized by monitoring the fault detection sensitivity at least two time points t ₁ 、t ₂ (where t is ₁ >t ₂ ) And increases by comparing it with a preset threshold. Also here, the variation may be due to a variation in load. Thus, in general, to avoid false fault detection, the fault detection may not be detectedThe settings are as sensitive as the zero sequence voltage based method.

The pick-up of the ground fault detection based on three times the negative sequence current magnitude should preferably be delayed so that transients that produce a negative sequence current are filtered out. This in effect means that the overcurrent condition should be active for at least a certain duration without a temporary drop in order to provide a final ground fault detection. In this condition, only a permanent source of negative sequence current, i.e. a single phase-to-earth fault, should preferably be detected as an earth fault. The detection of a relative earth fault based on a triple negative sequence current may in particular be used for a coarse detection of a relatively high current continuous earth fault.

According to an embodiment, in a compensated network, the 3I2_ start _ threshold is not predetermined, but can be determined in real time. This can be performed, for example, by the crowbar coil regulator or another unit or system connected thereto. The total system damping in amperes (Id) is determined and has a set detuning value in amperes (Iv). From these values, the desired ground fault current may be determined as: ief _ comp = abs (Id + j × Iv). In case the desired ground fault current is determined, then the 3I2_ start _ threshold may be automatically determined in real time as:

3I2_ start _ threshold ≦ Ief _ comp _ q1, where q1<1 is a user-defined safety margin. This embodiment may allow for an increased sensitivity for ground fault detection.

According to an embodiment, the detection 100 of the relative earth fault additionally or alternatively comprises: for triple zero-sequence current amplitudes (fundamental frequency only, fundamental frequency plus harmonic or harmonic only) or at least two times t ₁ 、t ₂ (where t is ₁ >t ₂ ) Monitoring of changes in amplitude therebetween; and comparing it with a preset threshold:

(with settable pickup delay), or

(with a settable pick-up delay),

or alternatively

(with a settable pick-up delay),

wherein the content of the first and second substances,

n =1,m =1, or

n =1 and m = any integer greater than 1, or

n = any integer greater than 1, and m is any integer greater than n,

wherein the indices t1 and t2 are associated with different moments (t 1> t 2)

And is wherein the content of the first and second substances,

is based on the theory of symmetrical components at frequency f _n * n (n = any integer greater than or equal to 1) calculated zero sequence current components.

This fault detection method may only be effective when the three zero sequence currents in the faulty feeder and the normal feeder have significantly different values. This condition may only be valid in ungrounded networks where the ground fault current produced by the protected feeder is significantly less than the total uncompensated ground fault current of the network.

According to an embodiment, an estimate of the earth fault current through the measurement point at the point of the detected phase-to-earth fault may be determined based on the faulty phase of the three-phase grid and the determined phase currents of the three phases, such that the estimate is based on the negative sequence current component. An example of this embodiment is the following formula 3b. According to a further embodiment, an estimate of the earth fault current through the measurement point at the point of the detected phase-to-earth fault may be determined based on the faulty phase of the three-phase power network and the determined change in the phase currents of the three phases. Examples of this embodiment include the following formulas 3a, 4a, and 5a to 5c.

According to an embodiment, the determination of the change of phase currents of the three phases of the three-phase electrical network comprises determining, for each of the three phases A, B, C of the three-phase electrical network 50, a difference between a fundamental frequency component of the phase currents during the phase-to-ground fault and a fundamental frequency component of the phase currents before the phase-to-ground fault.

According to an embodiment, the determination of the change of the phase currents of the three phases of the three-phase electrical network additionally or alternatively comprises determining for each of the three phases A, B, C of the three-phase electrical network and for at least one harmonic frequency that is an integer multiple of the fundamental frequency of the phase currents a difference between the harmonic frequency components of the phase currents during the phase-to-ground fault and the harmonic frequency components of the phase currents before the phase-to-ground fault. Thus, only the fundamental frequency component, only one or more harmonic frequency components, or both the fundamental frequency component and the one or more harmonic frequency components of the phase current may be used to determine the change in phase current for the three phases of the three-phase electrical network.

An estimate of the ground fault current through the measurement point at the point of the detected phase-to-ground fault may be determined according to various embodiments described below, or a combination thereof. The phase current at the measurement point 40 is measured, including measuring the fundamental frequency f _n A component and/or one or more harmonic components (n =2, 3, 4, 5, … …). For example, in some systems, in addition to the fundamental component, for example, the fifth and seventh harmonic components may actually dominate the ground fault current and may be included in the measurement. If the magnitude of any one or more of the harmonic components is sufficient for accurate measurement, it may be included in the measurement. The minimum value of such harmonic component amplitude may be a predetermined value and it may be determined, for example, by the hardware used and the accuracy of the applied measurements.

For at frequency n f _n (wherein f) _n Is fundamental system frequency) can be written as:

is at a frequency n f _n The current phasor of the lower phase a,

is at a frequency n f _n The current phasor of the lower phase B,

is at a frequency n f _n The phase C current phasor of the lower phase,

wherein n = any integer greater than or equal to 1

The "delta" or amount of change to each of the phase current phasors during a detected phase-to-earth fault may preferably be derived in substantially real time (in later calculations, if desired) according to the following equation:

where the subscript tF relates to the time instant during the detected ground fault and tP relates to the time instant before the detected ground fault (tF > tP).

Using the measurable change due to the ground fault at the phase current phasors enables an accurate estimation of the ground fault current.

According to an embodiment, an estimate of the ground fault current through the measurement point at the point of the detected phase-to-ground fault may be determined 110 by using any of the following equations 3a, 3b, 4a, and 5a to 5c, which may be evaluated individually or together. The A-B-C phase rotation is assumed in the exemplary formula:

equation 3a (based on the change in the negative-sequence current component due to the ground fault):

failure of phase a to ground:

phase B fault to ground:

phase C to ground fault:

equation 3b (based on the negative sequence current component during the fault without pre-fault data): failure of phase a to ground:

phase B fault to ground:

phase C to ground fault:

equation 4a (based on the change in the positive sequence current component due to the ground fault):

failure of phase a to ground:

phase B fault to ground:

phase C to ground fault:

equations 5a to 5c (based on the change in phase current due to the ground fault):

failure of phase a to ground:

or

Or

Phase B fault to ground:

or

Or alternatively

Phase C to ground fault:

or

Or

Wherein the content of the first and second substances,

is to assume phase A to ground fault at frequency n f _n Estimation of ground fault current

Is to assume phase B to earth fault at frequency n f _n Estimation of ground fault current

Is to assume a phase C fault to earth at a frequency n f _n Estimation of ground fault current

Is to assume phase A to ground fault at frequency n f _n Negative sequence current component of

Is to assume phase B to earth fault at frequency n f _n Negative sequence current component of

Is to assume a phase C fault to earth at a frequency n f _n Negative sequence current component of

Is to assume that the phase A fails to earth at a frequency n x f due to an earth fault _n Change in the negative-sequence current component below.

Is due to a ground fault at frequency n f when phase B is assumed to be faulty to ground _n Change in the negative-sequence current component below.

Is due to a ground fault at frequency n f when phase C is assumed to be faulty to ground _n Change in the negative-sequence current component below.

Is to assume phase A to ground fault at frequency n f _n Lower positive sequence current component

Is to assume phase B to earth fault at frequency n f _n Lower positive sequence current component

Is to assume a phase C fault to earth at a frequency n f _n Lower positive sequence current component

Is due to a ground fault at frequency n f when phase A is assumed to be faulty to ground _n Change of the lower positive sequence current component.

Is falsePhasing the frequency n x f due to ground fault when B is fault to ground _n Change of the lower positive sequence current component.

Is due to a ground fault at frequency n f when phase C is assumed to be faulty to ground _n Change of the lower positive sequence current component.

n =1, 2, 3, … … (integer)

According to an embodiment, equation 3b may be used, for example, in the event that the amount of current is not available before a fault. This may include, for example, special operating conditions, such as switching to a fault or during an auto-reclosing sequence. Furthermore, equation 3b may be used in network conditions where, for example, ground fault currents may be particularly high.

Determining an estimate of the ground fault current and determining the direction of the relative ground fault from the measurement point based on the determined estimate of the ground fault current may require detecting a faulty phase (phase a, phase B or phase C) of the electrical grid. Based on the detected faulty phase, the corresponding phase-specific earth-fault current phasor in terms of direction determination of the relative earth fault from the measurement point

Can be used as a ground fault current estimate

Phasor of (a).

The detection of such a faulty phase can be performed, for example, using any known method. Alternatively, the detection of a faulty phase may be performed according to one or more of the following embodiments:

when formula 5a1, formula 5b1, and formula 5c1 are applied:

the maximum value is indicative of a faulty phase,

when formula 5a2, formula 5b2, and formula 5c2 are applied:

if it is not

Providing the minimum value, the normal phase is A and B, and the fault phase is C

If it is not

Providing the minimum value, the normal phases are B and C, and the fault phase is A

If it is not

Providing the minimum value, the normal phase is C and A, and the fault phase is B

When formulas 5a3, 5b3, and 5c3 are applied:

if it is not

Providing the minimum value, the normal phase is A and C, and the fault phase is B

If it is not

Providing the minimum value, the normal phase is B and A, and the fault phase is C

If it is used

Providing the minimum value, the normal phase is C and B, and the fault phase is A

And/or using the phase-specific ground fault currents to estimate the real and zero sequence voltages (or derived ground fault power and ground fault admittance):

the maximum value indicates a faulty phase. Wherein the formula 5a1 to 5a3 or 5b1 to 5b3 or 5c1 to 5c3 are used

And/or by comparing the magnitude of the real part of the manipulated variable calculated using the fundamental zero-sequence voltage and the zero-sequence current, comprising:

wherein the content of the first and second substances,

is that

Phasors or

Phasor and

phase angle difference between phasors

Using the magnitude of the real part of the operational quantity calculated using the fundamental zero sequence voltage and the estimated ground fault current,

wherein the content of the first and second substances,

is that

Phasors or

Phasor and

phase angle difference between phasors

Therein, three equations are derived describing three possible phase-to-ground faults (phase a to ground, phase B to ground, and phase C to ground). For example, in the case of equation 6a6, which is based on an equation for ground fault admittance, then three equations describing three possible relative ground faults are:

the fault phase may be identified as the fault phase whose value is closest to the value calculated using the zero sequence voltage and the zero sequence current. Additionally, the sign of the real part of the operational quantity calculated with the fundamental zero sequence voltage and the zero sequence current should coincide with the sign of the real part of the phase-specific operational quantity calculated with the fundamental zero sequence voltage and the estimated ground fault current.

According to an embodiment, the estimation based on the determined ground fault current

And the determined zero sequence voltage or the determined change in the zero sequence voltage does not point out in the direction of the phase-to-earth fault from the measuring point 40The stator 120 may be according to a current-based approach, a power-based approach, an admittance-based approach, or a combination thereof. Examples of such methods are given below.

In the current-based method, the operation quantity is a ground fault current estimation

The phasor, for example, may be compared to a predetermined threshold value for its real or imaginary part or phase angle:

a)

b)

c) phi _ threshold 1<phi ⁿ <phi _ threshold 2

d) any combination of a) to c)

phi ⁿ During a ground fault

Phasors or

And

phase angle difference between phasors. Alternatively, phi ⁿ Is composed of

Phasors or

Phasor and

phase angle difference between phasors.

In a power-based approach, for example, the operational quantity is ground fault power

Phasors, whose amplitude, real or imaginary part or phase angle can be compared with a predetermined threshold:

a)

b)

c)

d) phi _ threshold 1<phi ⁿ <phi _ threshold 2

e) any combination of a) to d)

phi ⁿ During a ground fault

Phasors or

And

phase angle difference between phasors. Alternatively, phi ⁿ Is composed of

Phasors or

Phasor and

phase angle difference between phasors.

In admittance-based approaches, for example, the operational quantity is the ground fault admittance

a)

or

b)

c)

Or alternatively

d)

e)

Or

f)

g) phi _ threshold 1<phi ⁿ <phi _ threshold 2

h) any combination of a) to g)

phi ⁿ During a ground fault

Phasors or

And

phase angle difference between phasors. Alternatively, phi ⁿ Is composed of

Phasors or

Phasor and

phase angle difference between phasors.

According to an embodiment, in the compensation network, the current threshold, the power threshold or the admittance threshold are not predetermined, but may be determined in real time. This can be performed, for example, by the crowbar coil regulator or another unit or system connected thereto. The total system damping in amperes (Id) may be determined and may have a set detuning value in amperes (Iv).

From these values, the threshold value can be determined automatically in real time:

the real part _ threshold is less than or equal to Id q2,

the imaginary part _ threshold ≦ Iv × q3,

abs _ threshold ≦ Abs (Id + j Iv) q4

Wherein q2-q4<1 are user-defined safety margins.

According to an embodiment, the direction of the relative fault from the measurement point may then be determined 120 based on the comparison. According to an embodiment, the direction of the fault may be determined to be a first direction from the measurement point if the result of the comparison is true and/or may be determined to be a second direction from the measurement point if the result of the comparison is false. For example, in the case where the measurement point is located at the beginning of the line outlet, if the result of the comparison is true, it may be determined that the direction of the fault is toward the line outlet, and/or if the result of the comparison is false, it may be determined that the direction of the fault is away from the line outlet.

Alternatively or additionally, for example, any of the above criteria may be combined with a neutral point voltage condition and/or a residual current condition.

According to an embodiment, in order to obtain an accurate estimate of the ground fault current, the ground fault current estimates the phasor

Preferably at least the fundamental frequency component (n = 1).

According to an embodiment, the determination of the direction of the phase-to-ground fault from the measurement point can be performed by using one of a plurality of harmonic frequencies (n =2, 3, 4, 5, … …) in addition to the fundamental frequency. For example, if fifth and seventh harmonics are included, then in the current-based approach, the real or imaginary part or phase angle is compared to a predetermined threshold, and:

the operation amount based on the real part of the ground fault current can be calculated as follows:

phi ¹ is during ground fault

Phasors or

Phasor and

phase angle difference between phasors.

phi ⁵ Is during ground fault

Phasors or

Phasor and

phase angle difference between phasors.

phi ⁷ During ground fault

Phasors or

Phasor and

phase angle difference between phasors.

The operation amount based on the imaginary part of the ground fault current can be calculated as follows:

phi ¹ is during ground fault

Phasors or

Phasor and

phase angle difference between phasors.

phi ⁵ During ground fault

Phasors or

Phasor and

phase angle difference between phasors.

phi ⁷ Is during ground fault

Phasors or

Phasor and

phase angle difference between phasors.

In the above exemplary equation, the operation amount indicates the sum of the magnitude of the fundamental frequency and the magnitude of the one or more harmonic component currents. From a fault detection point of view, this may ensure and improve selectivity, since the amount of operation may become "boosted" due to the effects of possible harmonics. In addition, for example, the dependency may be enhanced because the determination of the direction of the phase-to-earth fault detection does not depend solely on the presence of harmonics or only on the fundamental frequency component.

According to an embodiment, a given harmonic may be included if its magnitude exceeds a predetermined measurable threshold. In practice this may be, for example, a few amperes. For example, if the magnitude of any harmonic component or harmonic components is sufficient for accurate measurement, it may be included in the determination. The minimum value of such harmonic component amplitude may be a predetermined value and may be determined, for example, by the hardware used and the accuracy of the applied measurement device (such as CT, VT and/or sensor).

According to an embodiment, after determining the ground fault current estimate, the determination of the direction to ground fault may be implemented by: independently include any current-based method, power-based method, or admittance-based method as previously described; or combining two or more of the methods. In addition, the conditions for determining the direction of the relative earth fault may include amplitude, real or imaginary part thresholds, or phase angle thresholds, either independently or by combining several methods. Such thresholds may be used independently or together, and they may depend on the grid or system in which any of the embodiments described herein are applied. For example, the determination of the direction of the relative ground fault may be made locally and/or centrally by comparing ground fault current estimates for some or all of the wire outlets at the substation.

According to an embodiment, the determined ground fault current estimation phasor may be converted to a ground fault admittance phasor using a zero sequence voltage according to the following equation:

admittance phasor for earth fault

Can be obtained by using

And

(or

) Is calculated. Alternatively, ground fault admittance

The accumulated phasor sum (CPS) method can be used during ground faults by using it as described in EP 2624397 A1

And

(or

) Is calculated. The sign of the calculated admittance is in this case

- "stabilized admittance", and it can be calculated according to equations 7c to 7 d:

due to a complete earth fault (fault resistance R) _F =0 Ω) is set in the case of (c),U ₀ is equal to the voltage U relative to the ground of the system _PE And thus can be converted by using a fixed scalar conversion factor U _PE Will be calculated

Value or

The value is converted from the admittance domain to the current domain. In addition, since the sign of the reactive component in the admittance domain is reversed, i.e. the capacitive susceptance is positive and the inductive susceptance is negative, the inductive susceptance is therefore negative

Or

Should be inverted, i.e. by applying the complex conjugate. Finally, the conversion from admittance domain to current domain becomes:

or

When fault resistance is involved in a fault, alsoAdmittance-based ground fault current estimation in the current domain

Or

Scaling is performed to match the actual value of the ground fault current. This may be done by estimating the ground fault current based on the admittance

Or

Multiplied by the measured relative zero sequence voltage amplitude

To realize that:

or

Marking

Or

An admittance-based ground fault current estimate in the current domain is represented, which is obtained by converting from the calculated admittance, taking into account the damping effect of the fault resistance. Alternative ground fault current estimation

Admittance based ground fault current estimation

Or

May be used in all calculations for the various embodiments described herein.

Such admittance-based equations may be used with CPS calculations have the benefit that they may provide a very stable estimate of the ground fault current regardless of fault-type-related oscillations in the measured current and/or voltage quantities (e.g., during a re-breakdown ground fault). This in turn enables reliable and accurate implementation and performance of various applications using ground fault current estimation.

According to an embodiment, all ground fault current estimates may be converted to ground fault power using the measured zero sequence voltage according to the following equation:

ground fault power

Can use

And

(or

) Is calculated.

Alternatively, ground fault power

Can be obtained by using cumulative phasors and methods as described in EP 2624397 A1

And

(or

) Is calculated.

Or

Then ground fault power

Can be calculated as:

or alternatively

For example, an operating quantity, such as current, admittance, or power, determined from the accumulated phasors, by using a Cumulative Phasor Sum (CPS) method and the determined zero-sequence voltage or a variation of the determined zero-sequence voltage, may be used for a current-based method, a power-based method, an admittance-based method, or a combination thereof.

Various embodiments described herein provide advantages in that: detection of a faulty feeder (or feeder line segment) and a normal feeder (or feeder line segment), for example based on an estimated ground fault current, increases the reliability of selective fault detection. For the calculation of the ground fault current in the compensation network, the following equation is valid (assuming complete symmetry of the system).

Neutral point voltage during single-phase earth fault:

attenuation of neutral point voltage and ground fault current due to fault resistance:

wherein the content of the first and second substances,

·U _PE is operating voltage [ V ] to ground]

·I _EFFd Is a capacitive earth fault current [ A ] generated by a protected feeder]

·I _EFNet Is an uncompensated capacitive earth fault current [ A ] generated by the network]

·d _Net Is a factor [ pu ] approximating the natural loss of the feeder/network]Typical values are from I _EFNet Or I _EFFd Between 0.01 … … 0.10.10

·I _Coil Is an inductive current [ A ] generated by an ASC (ASC, arc suppression coil)]Determined by the degree of tuning set

·d _Coil Is a factor [ pu ] approximating the ASC loss]Typical values are from I _Coil Between 0.01 … … 0.05.05

·I _Par Is an additional resistive current [ A ] at the level of the primary voltage generated by the parallel resistor of the arc suppression coil]

·R _F Is fault resistance ohm]

Earth fault current measured at the beginning of a normal feeder during a phase-to-earth fault:

earth fault current measured at the beginning of a faulty feeder during a phase-to-earth fault (fundamental frequency component):

or by rearranging the items:

the imaginary part of the denominator, term j (I) _EFNet -I _Coil ) Is the degree of detuning of the network in amperes. The real part of the denominator is equal to the total network loss: parallel resistors, losses in the coils and losses in the network.

The detection of faulty and normal feeders or feeder line segments based on an estimation of the ground fault current as proposed by various embodiments may have the following advantageous features, compared to residual current based methods, for example:

1. earth fault current measured in normal feeder during single-phase earth fault

Theoretically zero.

Thus, the magnitude of the total relative ground admittance of a normal feeder (feeder topology, connection state or share of underground cables) does not affect the ground fault current

And (6) estimating. This is advantageous, especially in case of normal feeders with large total relative admittance and with small losses (e.g. long cable feeders). In this case, the security of the protection can be increased.

2. Earth fault current measured at the beginning of a faulty feeder during a single-phase earth fault

The estimate corresponds better to the actual ground fault current flowing at the fault location.

Therefore, the ground fault current

The estimated magnitude is not affected, or at least less affected, by the contribution of the ground fault current generated by the faulty feeder itself (feeder topology, connection state or underground cable contribution).

From a protective point of view, these facts can bring the following benefits:

a) In the case where the network is operating in an under-compensated mode and the degree of under-compensation is equal to the uncompensated capacitive ground fault current produced by the protected feeder, the imaginary part of the residual current measured in that feeder during a ground fault condition

Theoretically equal to zero. However, the measured ground fault current estimate

Is theoretically equal to a magnitude determined by at least the degree of under-compensation. This means that the dependability of the protection depends not only on the resistive component of the estimated ground fault current (due to losses in the network and the coil itself), but also on the measurable reactive part of the ground fault current.

b) Additionally, for example, in addition to the fundamental frequency component (n = 1), the ground fault current estimation

May also include a harmonic component (n =2, 3, 4, … …) resulting in a higher measurable current at the faulty feeder during a ground fault in the presence of the harmonic component. The ability to include harmonic components to protect the operational quantities enhances the protection's dependability.

c) Ground fault current estimation

Unaffected or at least less affected by the protection feeder parameters (e.g. feeder topology, connection status or share of underground cable), but with ground fault current estimation

Mainly depending on the total network admittance. Thus, also in case the protected feeder has a high total relative admittance compared to the total relative admittance of the background network and the coil, such a ground fault inside the feeder may be dependently and selectively detectable.

d) For example, in a compensation network, the total network admittance value can be easily determined by the network parameters calculated by the coil controller. The total network admittance value can also be easily calculated using basic network parameters for any high impedance grounded network. This makes it very easy to set the criteria or even to make an automatic adjustment of the settings in real time using the parameters calculated by the coil controller.

e) In the case where the coil (ASC) produces a very high inductive component compared to the capacitive component produced by the network admittance (when the network is operating in a high overcompensation condition), the high detuning value of the ASC is proportional to the magnitude of the ground fault current estimate. Thus, the magnitude of the ground fault current estimate may be used as a selective indication of a high ground fault current condition in the network.

The various embodiments described herein can readily take into account such extreme and very challenging fault conditions.

3. In-ground fault current estimation

May not see oscillations after the fault. When the fault arc is extinguished, the ground fault current will become zero and will be unaffected or at least less affected by the post-fault oscillation than the zero sequence quantity. In this case, the security of the protection can be increased.

4. Estimation of earth-fault current due to start of measurement of normal feeder during single-phase earth fault

Theoretically zero and ground fault current estimation measured at the beginning of a faulty feeder during a single phase ground fault

Corresponds to the actual ground fault current flowing at the fault location and is therefore based on the ground fault current

The estimated ground fault detection method allows the amplitude alone to be used as a dependable amount of protection. Thus, ground fault protection in a high impedance grounded network can be selectively operated even in the absence of a directional determination, i.e. operation in no direction. This is especially not possible in compensation networks when traditional methods based on the amount of residual current have been used.

Thus, the operation of the protection may be estimated, for example, by simply monitoring the ground fault current at the protected feeder

Is implemented in terms of magnitude, real or imaginary part, or phase angle (free-standing application).

Ground fault current estimates between feeders at a substation may also be compared

And selecting one of the faults based on the comparison (centralized application).

To account for possible measurement inaccuracies, earth-fault current estimation

Should preferably exceed some preset minimum value.

Thus, selective protection is possible even without the parallel resistor of the Arc Suppression Coil (ASC), but this may require that the arc suppression coil not be tuned to full resonance, and thus, for example, slightly over-compensated or under-compensated operation.

5. Since the ground fault current magnitude is proportional to the dangerous voltage at the fault location, adaptive protection can be automatically performed based on the estimated magnitude of the ground fault current.

Typically, this may mean that the operating speed of the protection may be increased when the ground fault current increases and the operating time may be delayed in case the ground fault current is small, for example.

6. Furthermore, the set criteria for ground fault protection according to various embodiments may be much simpler, since e.g. the ground fault current estimation may be based directly on the expected ground fault current

To determine the set threshold.

In particular, in a compensation network, the setting will be very easy, since for example the set values may be directly based on network parameters (e.g. detuning values and network damping values) calculated by the coil controller.

7. Ground fault current estimation

Can be further divided into magnitude, real and imaginary parts. It is also possible to multiply or divide with zero sequence voltage, resulting in protection based on ground fault power or based on ground fault admittance, for example.

The proposed protection method and apparatus according to various embodiments also allow harmonic components to be included in the estimation of the earth fault current, making the solution much more accurate in modern power systems where such harmonics are likely to occur in practice.

For example, the proposed protection method and apparatus according to various embodiments may be a protection function independent or complementary to existing residual voltage and current based ground fault protection.

A device according to any one or combination of the above embodiments may be implemented as a single unit or as two or more units configured to implement the functionality of the various embodiments. Herein, the term "unit" generally refers to a physical or logical entity, such as a physical device or a part of a physical device or a software routine. One or more of these units may reside in the protective relay unit 41, for example.

The device for implementing the described functionality according to any of the embodiments may be implemented, for example, at least partly by means of one or more computers or corresponding Digital Signal Processing (DSP) devices provided with suitable software. Such a computer or digital signal processing apparatus preferably includes at least a work memory (RAM) providing a storage area for arithmetic operations and a Central Processing Unit (CPU) such as a general-purpose digital signal processor. The CPU may include a set of registers, an arithmetic logic unit, and a control unit. The CPU control unit is controlled by a series of program instructions transferred from the RAM to the CPU. The CPU control unit may contain a plurality of microinstructions for basic operations. The implementation of the microinstructions may vary depending on the CPU design. The program instructions may be encoded by a programming language, which may be a high-level programming language such as C, java or a low-level programming language such as a machine language or assembler. The computer may also have an operating system that may provide system services to computer programs written in the program instructions. A computer or other apparatus implementing the invention or part thereof may also include suitable input means for receiving, for example, measurement data and/or control data, and output means for outputting, for example, control data or other data. Application specific integrated circuits or discrete electrical components and devices for implementing the functions according to any of the embodiments may also be used.

The present invention may be implemented with existing system elements, such as various protective relays or similar devices, or by using individual dedicated elements or devices in a centralized or distributed manner. Current protection devices for electrical systems, such as protection relays, may include processors and memory that may be used in the functionality according to the various embodiments described herein. Thus, all modifications and configurations required for implementing an embodiment in an existing electrical system component may be performed as software routines, which may be implemented as added or updated software routines. If at least part of the functionality of the invention is implemented by software, such software may be provided as a computer program product comprising computer program code which, when run on a computer, causes the computer or a corresponding appliance to perform the functionality according to embodiments as described herein. Such computer program code may be stored or typically embodied on a computer readable medium such as a suitable memory, e.g. flash memory or optical memory, from which it can be loaded into one or more units executing the program code. Furthermore, such computer program code implementing the invention may for example be loaded via a suitable data network to one or more units executing the computer program code and may replace or update program code that may be present.

It will be obvious to a person skilled in the art that as technology advances, the inventive concept can be implemented in various ways. The invention and its embodiments are not limited to the examples described above but may vary within the scope of the claims.

Claims

1. A method for use in ground fault protection in a three-phase electrical network, comprising:

detecting (100) a relative earth fault in the three-phase power network (50);

determining (110), for each of three phases (A, B, C) of the three-phase electrical network, phase currents during the detected phase-to-earth fault or changes in the phase currents due to the detected phase-to-earth fault at a measurement point (40) in the three-phase electrical network (50);

detecting a faulted phase of the three-phase power grid;

determining (110) an estimate of a ground fault current through the measurement point (40) at the point of the detected phase-to-ground fault (60) based on the faulty phase of the three-phase grid and the determined phase currents of the three phases or based on the faulty phase of the three-phase grid and the determined variations of the phase currents of the three phases, wherein the estimate is based on a negative sequence current component;

determining a zero sequence voltage of the three-phase power network during the detected phase-to-earth fault or a change in the zero sequence voltage of the three-phase power network due to the detected phase-to-earth fault; and

determining a direction of the phase-to-earth fault from the measurement point (40) based on the determined estimate of the earth fault current and the determined zero sequence voltage or based on the determined estimate of the earth fault current and the determined variation of the zero sequence voltage.

2. The method according to claim 1, comprising operating the earth fault protection in the three-phase power network based on the determined direction of the relative earth fault from the measurement point (40).

3. The method of claim 2, wherein the operation of ground fault protection comprises:

tripping and/or preventing tripping of one or more switching devices in the three-phase electrical network; and/or

Disconnecting or limiting the earth fault current of the detected phase-to-earth fault in the three-phase electrical network (50).

4. The method according to any one of claims 1 to 3, wherein the determination of the change of the phase currents of the three phases of the three-phase electrical network comprises:

for each of the three phases (A, B, C) of the three-phase electrical network, a difference between a fundamental frequency component of the phase current during the phase-to-earth fault and a fundamental frequency component of the phase current before the phase-to-earth fault is determined.

5. The method according to any one of claims 1 to 3, wherein the determination of the change of the phase currents of the three phases (A, B, C) of the three-phase electrical network (50) comprises:

determining, for each of the three phases (A, B, C) of the three-phase electrical network (50) and for at least one harmonic frequency that is an integer multiple of a fundamental frequency of the phase current, a difference between a harmonic frequency component of the phase current during the phase-to-earth fault and a harmonic frequency component of the phase current before the phase-to-earth fault.

6. The method of claim 4, wherein the estimate of the ground fault current through the measurement point (40) at the point of the detected phase-to-ground fault (60) is determined by using at least one of the following equations: an equation based on the change of the negative sequence current component, an equation based on the change of the positive sequence current component, and an equation based on the change of the phase current.

7. A method according to any of claims 1-3, wherein an estimate of the earth fault current through the measurement point (40) at the point of the detected phase-to-earth fault (60) is determined during the fault by using a formula based on the negative sequence current component.

8. A computer-readable medium storing computer program code, wherein execution of the computer program code in a computer causes the computer to perform the steps of the method according to any one of claims 1 to 7.

9. An apparatus for use in ground fault protection in a three-phase electrical network, comprising:

-means configured to monitor phase currents of the three-phase electrical network (50);

-means configured to detect a phase-to-earth fault in the three-phase electrical network (50);

means configured to determine, for each of three phases (A, B, C) of the three-phase electrical network, at a measurement point of the three-phase electrical network (50), a phase current during the detected phase-to-earth fault or a change in the phase current due to the detected phase-to-earth fault;

means configured to detect a faulty phase of the three-phase electrical network;

means configured to determine an estimate of a ground fault current through the measurement point at the point of the detected phase-to-ground fault (60) based on the faulty phase of the three-phase electrical network (50) and the determined phase currents of the three phases or based on a change in the faulty phase of the three-phase electrical network and the determined phase currents of the three phases, such that the estimate is based on a negative sequence current component;

means configured to determine a zero sequence voltage of the three-phase electrical network during the detected phase-to-earth fault or a change in the zero sequence voltage of the three-phase electrical network due to the detected phase-to-earth fault; and

-means configured to determine a direction of the relative earth fault from the measurement point (40) based on the determined estimate of the earth fault current and the determined zero sequence voltage or based on the determined estimate of the earth fault current and the determined change in the zero sequence voltage.

10. The apparatus of claim 9, comprising:

means configured to operate the ground fault protection in the three-phase electrical network based on the determined direction of the relative ground fault from the measurement point (40).

11. The apparatus of claim 10, wherein the device configured to operate the ground fault protection comprises:

means configured to trip and/or prevent tripping of one or more switching devices in the three-phase electrical network; and/or

Means configured to disconnect or limit the earth fault current of the detected phase-to-earth fault in the three-phase electrical network (50).

12. The apparatus according to any of claims 9 to 11, wherein the device configured to determine, for each of three phases (a, B, C) of the three-phase electrical network, a change in the phase current due to the detected phase-to-earth fault at the measurement point is configured to:

13. The apparatus of any of claims 9 to 11, wherein the device configured to determine, for each of the three phases (a, B, C) of the three-phase electrical network, a change in the phase currents due to the detected phase-to-earth fault at the measurement point is configured to:

14. The apparatus of claim 12, wherein the means configured to determine an estimate of the ground fault current is configured to determine an estimate of the ground fault current through the measurement point (40) at the point of the detected phase-to-ground fault (60) by using at least one of: an equation based on the change of the negative sequence current component, an equation based on the change of the positive sequence current component, and an equation based on the change of the phase current.

15. The apparatus of any of claims 9 to 11, wherein the device configured to determine the estimate of the ground fault current is configured to: determining an estimate of the ground fault current through the measurement point (40) at the point of the detected phase-to-ground fault (60) during a fault by using an equation based on the negative sequence current component.