WO1996032652A1

WO1996032652A1 - Distance measurement process

Info

Publication number: WO1996032652A1
Application number: PCT/DE1996/000628
Authority: WO
Inventors: Andreas Jurisch
Original assignee: Siemens Aktiengesellschaft
Priority date: 1995-04-13
Filing date: 1996-04-03
Publication date: 1996-10-17
Also published as: DE19514698C1; EP0820600A1

Abstract

A distance protection measurement process is disclosed for multiple-phase electric energy transmission lines. Voltage and current in the faulty phase conductor are sensed, digitised and evaluated in linear phase response, non-recursive digital filters (FIR filters) of a filter unit. The weighting factors of the FIR filters are freely predetermined and errors are corrected by means of a correction factor. The distance between the faulty area and impedance measurement values that indicate the measurement site are derived in a computer from the output values of the filter unit. In order to measure distances with accuracy even in the case of monopolar short-circuits to ground, a total current (IOFA) that corresponds to the sum of the currents in the phase conductors of the energy transmission line is sensed, digitised and evaluated in additional FIR filters (8, 9) of the filter unit (1), forming output values (mk, nk). The computer (10) calculates four auxiliary values with which it calculates, together with the output values (yk, mk, nk, wk, vk) of the filter unit (1), a length factor (m) and a resistance (Rf) that is proportional to the resistance of the faulty area. By multiplying the length factor (m) by the kilometric resistance (R'1) of the associated system and by adding the resistance value (Rf) and multiplying the kilometric reactance (L'1) of the associated system by the length factor (m), the measurement impedance (R, X) that characterises the distance from the faulty area is calculated.

Description

description

Method for performing a distance measurement If a method for performing a distance measurement on a multi-phase electrical power transmission line is carried out digitally using a computer, measurement errors result that are specific to the algorithm. In addition, there are further measurement errors due to the assumption that the measured equation and the following equation (1)

Writable loop reactance X as an imaginary part of the quotient from the loop voltage U _Schl and the loop current

l _Schl the error distance is directly proportional. This assumption only applies if the energy transmission line is open at the other end. As a rule, the load on the line at the time of the fault occurrence and the supply of the fault current from the other end of the energy transmission line result in additional measurement errors which can reach the order of magnitude of 50% and more, particularly in high-voltage networks. Figure 1 illustrates these measurement errors in principle using an equivalent circuit diagram for a simple energy transmission line fed from two sides.

As can be seen in detail in FIG. 1, in the assumed case an electrical energy transmission line with two conductors L ₁ and L _{2 is} fed from two supply points Ua and Ub. A load current I _{L flows} . The individual impedances of the system shown are a further one due to an equivalent impedance Z _1A for the area from the feed point Ua to a measurement point A close to the feed point Substitute impedance mZ _1L for the part of the energy transmission line from measuring point A to an assumed fault location F, a further substitute impedance (1-m) Z _1L for the rest of the energy transmission line and an additional substitute impedance 1 _{1B of} the other supply point Ub; the equivalent impedance of the

Fault point F itself is labeled 3.R _f and the current through the fault point is l _F.

The voltage U _RA measured at the measuring point A in the event of a short circuit at the fault location F is composed of several partial voltages. Thus, the fault current I _FA emitted by the supply point Ua in the event of a short circuit generates a voltage drop across the conductor loop to be measured. Furthermore, the load current I _L generates a voltage drop at the equivalent impedance mZ _1L for the line section between the measuring point A and the fault point F. A further partial voltage to be taken into account arises from the fault current I _FB fed from the supply point UB from the other end of the line via the fault resistor

3R _f . The latter two partial voltage drops cause the measurement errors mentioned. In order to be able to correct these measurement errors exactly, they must be quantified precisely beforehand.

For a precise distance measurement with a distance protection arrangement, a load current compensation is required, which until now has only been common with fault locations. In the book by H. Opperskalski

"Behavior of impedance-determining distance protection algorithms", progress reports VDI series 6 power generation No. 256, pages 45 to 47 describes a method that uses an iteration method to calculate the fault location. Such methods have no defined runtime behavior, since the computing time required to carry out the method depends on the unpredictable number of iteration cycles required. In principle, such a method is not suitable for a real-time application, as is the case with a distance measurement in A distance protection arrangement is required in order to be able to switch off the faulty line section in the shortest possible time in the event of a short circuit. Another iteration method for determining a fault location is described in British Patent Application GB 2 036 478 A. In this method, among other things, a current and a voltage are calculated at the fault location, and the phase relationship of the two variables to one another is then determined; If there is a phase deviation, a different voltage is calculated at the fault location and the phase position of the calculated current is determined again. If phase correspondence is finally reached, the location of the fault is inferred from the variables taken into account.

Furthermore, in the European patent application

EP 0 106 790 AI describes a method for localizing a fault location on an overhead line, in which the calculation of complex current and voltage pointers is carried out in a first method step. In a second method step, a quadratic equation is solved using the complex impedances of the line itself, which are assumed to be known, and the feed-in impedances of both line ends, as well as the type of error. Because of the known complex feed impedances, this method is only suitable for use in fault locations, since the feed impedances in a typical network depend on the switching status of the network and data transmission is therefore required to provide this information in the protective device. A distance protection arrangement must, however, be functional independently of such data connections.

Furthermore, the arithmetic effort for calculating the fault location using this known method is lower than with the iteration method discussed above, but still very high.

A method for testing arrangements is also known (European patent specification EP 0 284 546 B1), which can be used to precisely determine the fault location on an electrical power transmission line. In this known method, the current and voltage of the energy transmission line or quantities derived therefrom are processed in a filter unit with non-recursive digital filters (FIR filters); In a subordinate arithmetic unit, the values indicating the location of the error are calculated therefrom after error correction. In order to be able to work with this method precisely and in real time as part of a distance protection arrangement, a relatively powerful and thus relatively expensive computer must be used.

The invention is based on the method last dealt with above, that is to say relates to a method for carrying out a distance measurement on a multi-phase electrical power transmission line, in which the voltage on a faulty phase conductor is detected, digitized and in a linear-phase, non-recursive digital filter (FIR filter). of a first type (with weighting factors g _i ) of a filter unit, which detects current in the faulty phase conductor,

is digitized and evaluated on the one hand in a further FIR filter of the first type and on the other hand in a FIR filter of a second type (with weight factors f _i ) of the filter unit, the weight factors being freely specified and an error correction being carried out by means of a correction factor which is used as Quotient is formed from the amplitude responses of the FIR filters of the first and second type, and from the output variables of the filter unit in a computing unit, the respective distance of the fault location from a measuring location Impedance measures are determined, and the

Task to develop this known method so that it is particularly well suited for distance measurement. To achieve this object, according to the invention, the sum of the is used for distance measurement in the case of single-pole earth faults

Currents in the phase conductors of the energy transmission line are detected, digitized and evaluated on the one hand in an additional FIR filter of the first type and on the other hand in an additional filter of the second type of the filter unit, each with the formation of an output variable; a first auxiliary variable is formed in the arithmetic unit from the output variable of the further FIR filter of the first type by multiplication by the kilometric ohmic resistance of the co-system of the energy transmission line, and in the arithmetic unit the output variable of the one FIR filter of the second type is also formed by multiplication with the

kilometric inductance of the co-system of the energy transmission line formed a second auxiliary variable; in the

Computing unit is also obtained from the output variable of the additional FIR filter of the first type by multiplying by the difference between the kilometric ohmic resistance of the zero system and the co-system of the energy transmission line and a third auxiliary variable from the output variable of the additional FIR filter of the second type

Multiplication with the difference between the kilometric inductance of the zero system and the co-system of the energy transmission line forms a fourth auxiliary variable; in the arithmetic unit, the output variables of the FIR filter and the auxiliary variables are converted into a length factor and a resistance at the

Fault location proportional resistance value calculated and by multiplying the length factor by the kilometric

Resistance of the co-system and addition of the resistance value as well as by multiplying the kilometric reactance of the Mitsystems with the length factor formed the impedance characterizing the distance of the fault location.

An essential advantage of the method according to the invention is that it only has two additional FIR filters in the filter unit and one in comparison to the known method with regard to a fault location determination

Insignificantly increased computing effort in the computing unit allows an accurate distance measurement in real time with single-pole earth faults through load current compensation. This is done using only the measured variables available at the measuring location and with knowledge and taking into account the kilometric resistance and reactance of the co-system and the counter-system of the energy transmission line to be protected. This method is characterized by a precisely defined runtime and a particularly low arithmetic effort and is therefore particularly suitable for real-time applications.

In order to ensure the accuracy of the distance measurement in the method according to the invention even in the case of a further energy transmission line parallel to the energy transmission line to be monitored, in an advantageous development of the method according to the invention, a zero current corresponding to the sum of the currents in the phase conductors of this energy transmission line is present in the presence of a parallel multiphase electrical energy transmission line recorded and digitized and the real part of the summed up

Current at the input of the additional FIR filter of the first type is added to the digitized total current and the imaginary part of the totalized current at the input of the additional FIR filter of the second type is added to the digitized total current.

A distance measurement must not only be carried out quickly in a distance protection device, but it must also be accurate and reliable so that the device does not trigger and thus switch off the energy transmission line to be monitored due to an inaccurate measurement. For this reason, distance protection devices work with so-called repeat measurement; however, additional time is required for this, even if - as will be shown later - the distance measurement itself was accurate. In this regard, a further development of the method according to the invention is advantageous, in which a further distance measurement is carried out in parallel, in that the voltage on the faulty phase conductor in a supplementary FIR filter of a third type (with weighting factors h _i ) in the filter unit to form an output variable is evaluated, the current in the defective phase conductor is evaluated in a further supplementary FIR filter of the third type to form an output auxiliary variable, the total current in an additional supplementary FIR filter of the third type is evaluated to form an additional output auxiliary variable in which A first additional auxiliary variable is formed from the output auxiliary variable of the further supplementary FIR filter of the third type by multiplication with the kilometric resistance of the co-system of the energy transmission line, and furthermore from the output variable of the wide one Ren FIR filter of the first type is formed by multiplication with the kilometric inductance of the co-system of the energy transmission line, a second additional variable is formed in the arithmetic unit from the output auxiliary size of the additional supplementary FIR filter of the first type by multiplication by the difference kilometric ohmic resistance of the zero system and the co-system of the energy transmission line, a third additional variable is obtained in the computing unit, moreover, from the output variable of the additional supplementary FIR filter of the third type by multiplication with the difference from the kilometric inductance of the zero system and the co-system of the Power transmission line one fourth additional variable is formed, in the arithmetic unit a length factor and a resistance value proportional to the resistance at the fault location is calculated from the output variables of the FIR filter and the additional variables, in the arithmetic unit by multiplying the length factor by the kilometric reactance of the co-system with the length factor one Distance of the fault location characteristic comparison impedance is formed, and the distance measurement is considered to be sufficiently accurate if the difference between the measurement impedance and the comparison impedance remains within a predetermined size.

To further explain the invention is in

FIG. 2 shows a component network of an electrical power transmission line to be monitored in the event of a single-pole earth fault, in

Figure 3 in the form of a block diagram an embodiment of an arrangement for performing the method according to the invention, in

Figure 4 shows another embodiment of an arrangement for

Implementation of the procedure and in

Figure 5 shows an additional embodiment.

For a single-pole earth fault on a three-phase power transmission line, the equivalent circuit diagram shown in FIG. 2 applies, in which I denotes the co-system, II the opposite system and III the zero system. 2 thus shows the relationships on the multiphase power transmission line in symmetrical components in a representation, which e.g. the book by R. Roeper "Short-circuit currents in three-phase networks", 1984, pages 48 to 51 can be found.

From the two feed points Ua and Ub, a load current I _{1FA is} generated only by the co-system; in addition, a fault current l _F arises. The distribution of the fault current I _F among the individual Parts I to III of the component network are calculated as follows using the current distribution factors c ₀ and

I _{0FA is} a part of the fault current I _F and I _0FB , which denotes the further part of this fault current; the equivalent impedances in the three parts I to III of the component network are defined in accordance with FIG. 1. The current I _0FA corresponds to the sum of the currents in the individual phase conductors of the energy transmission line to be monitored. If one sets up the mesh equation for the mesh entered in FIG. 2, one obtains after the back transformation into natural components:

The influences of the load current I _FA and the feed from the feed point UB are already taken into account in this formula. However, this equation still contains the quantity I _{t that} cannot be measured at point A. By inserting the current distribution factor co for the zero system, you get:

Assuming that the phase angles of the two currents l _0FA and l _0FB do not differ from one another, the fault resistance R _f and the current division factor c ₀ can be combined to form a fictitious fault resistance R _cf :

This equation is linear and, in addition to the line constants and the signals measurable at the installation site, only contains the two unknowns m and R _cf. After transforming the above equation into the time domain, we get:

The parameters m and R _cf are therefore to be determined, which is done with the method known from the above-mentioned European patent 0 284 546 B1. For this purpose, the variables u _RA , i _OFA and i _{FA are} evaluated after standardization in a filter unit 1 according to FIG. 3. Such an evaluation is carried out using convolution operations (symbolically represented with * in the block diagram). For this purpose, the normalized voltage u _RA is fed to a linear-phase, non-recursive digital filter, that is to say an FIR filter 3, via an analog-digital converter 2, which converts the voltage u _RA into a number sequence u _k after sampling with a correspondingly selected sampling time Ta. This FIR filter 3 belongs to a first filter type and has a symmetrical weight factor distribution g _i = g _ni ; n denotes the degree of the counter. At the output of the FIR filter 3 there is a sequence y _k , the mapping rule of which is:

Furthermore, after corresponding scanning in a further analog-digital converter 4, the standardized variable i _{FA is} converted and the resulting values x _{k are} fed to a further FIR filter 5, which also belongs to the first filter type and whose weight factor distribution is identical to that of the FIR Filters 3; At the output of the further filter 5, a sequence w _{k is} generated, which is described with:

In addition, the values x _{k are} supplied to a further FIR filter 6, which belongs to a second type whose weight factor distribution is: f _i = - f _ni (10)

At the output of this FIR filter 6 there is a sequence v _k , the mapping rule of which is (again with "x" for current samples):

After normalization, the total current i _OFA is supplied to an additional analog-digital converter 7, which outputs a sequence of numbers i _ok at the output. This sequence of numbers is folded in an additional FIR filter 8 of the first type, whereby an output variable m _{k is} formed at the output of this filter. In addition, an additional output variable n _{k is} generated in an additional FIR filter 9 of the second type.

The basic principle is that the FIR filters used satisfy the relationship

F (p) = p. G (p). (12)

Equation (6) for is equivalent to the procedure in the measurement method according to European patent specification 0 284 546 two different times T ₁ and T ₂ set up and resolved according to the two unknown quantities m and R _cf. The following specification for m and R _{cf is obtained} :

with k ₁ = R ₁ * G * I _FA1 - (R ₀ -R ₁ ) * G * I _OFA1 + L ₁ * F * I _FA1 - (L ₀ -L ₁ ) F * I _OFA1 (15) k ₂ = R ₁ * G * I _FA2 - (R ₀ -R ₁ ) G * * I _OFA1 + L ₁ * F * I _FA2 - (L ₀ -L ₁ ) F * I _OFA1 (16)

R ₁ · G * l _{FA1.2 denotes} a first auxiliary variable Hl,

L ₁ · F * l _FA1,2 a second auxiliary variable H2,

(R ₀ -R ₁ ) · G * l _OFA1,2 a third auxiliary quantity H3 and

(L ₀ -L ₁ ) · E * l _{O FA1.2} a fourth auxiliary variable H4.

The index numbers "1" and "2" identify the values of I _FA and I _OFA sampled at different sampling times. The quantities R and X required for the polygon arrangement are obtained from these calculation results. The actual error resistance is not reconstructed from the calculated virtual error resistance R _cf. The actual fault resistance R is calculated using the following formula:

R _f = c ₀ · R _cf

From the ratio of the current I _{j determining} the voltage drop across Rf to the measurable residual current component I _OFA and its

Angle, the following relationship can be derived using the current divider rule:

The angle β normally has a very small value in energy systems. A range of 0..6 ° is sometimes specified. It can therefore be assumed that the correction of the direct measurement will have relatively little influence on the determined reactance X. Since it is relatively easy to set an arc reserve, the virtual error resistance R _{cf is also} not corrected. With these requirements, the sizes used for polygon classification are calculated according to the following rule:

It is with the kilometric reactance and with

denotes the kilometric resistance of the energy transmission line to be monitored.

This type of calculation of the quantities used for the polygon arrangement has the advantage that no parameters for describing the pre-impedances of the line to be protected are necessary.

In order to take into account an inductive coupling through the zero current of a neighboring line parallel to the monitoring energy transmission line, a method is used which can be illustrated by the block diagram shown in FIG. 4. In this method, a zero current ioMA of a neighboring system (not shown) (sum of the currents in the phase conductors of the neighboring system) is supplied after standardization to an additional analog-to-digital converter 12, which is followed by an arithmetic logic unit 13. This arithmetic unit generates an additional variable ZG1 at its one output AI, which corresponds to the real part Re {K _OM ) · l _OAM ; A further additional variable ZG2 is formed at the output A2 corresponds to the imaginary part lm {K _OM ] · l _OAM . The sums of these portions with the size x _{k are} formed in subordinate summers 14 and 15.

Inductive coupling through the zero-sequence current of the neighboring system is taken into account by means of the complex correction factor k _OM . The real and imaginary part of the complex factor k _OM each represents a parameter of the protective device. These parameters reflect the ratio of the intersystem coupling of the two subsystems of the double line to the line impedance of the line to be protected.

In the computing unit 10 of the influence of a neighboring line is taken into account in that in the above equations (15) and (16) the terms are F * G * l _OA1 and _OA1 l replaced by the post-standing Asdrücke.

F * I _OΛ1 = F * (l _OA1 + Im { _kOM } · l _OAM1 ) F * I _OA2 = F * (l _OA2 + Im {k _OM } · I _{OMA 2} )

G * I _OA1 = - G * (l _OA1 + Re {k _OM } · I _OMA1 ) G * I _OA2 = G * (l _OA2 + Re {k _OM } · I _OMA2 )

In the exemplary embodiment according to FIG. 5, components that correspond to the exemplary embodiment according to FIG. 3 have been provided with the same reference symbols. In contrast to the exemplary embodiment according to FIG. 3, a filter device 16 is constructed differently here in that, in addition to the FIR filters 3, 5, 6, 8 and 9 according to the exemplary embodiment according to FIG. 3, it has a supplementary FIR filter 17 of a third type with weight factors hi, in which the voltage u _{RA is} evaluated by a folding operation; at the output of the additional FIR filter 17 there is an output auxiliary variable o _k . Furthermore, a further supplementary FIR filter 18 of the third type is arranged in the filter unit 16, in which the current in the faulty phase conductor of the energy transmission line to be monitored is evaluated; On the output side, a further auxiliary output variable p _k occurs at this FIR filter 18. Finally is the filter unit 16 is also equipped with an additional additional FIR filter 19 of the third type by evaluating the total current i _OFA . An additional auxiliary output variable r _k results at the output of this additional FIR filter 19.

With regard to the design of the additional FIR filters 17, 18 and 19 of the third type, it should be pointed out that the FIR filters G (jΩ), F (jΩ) and H (jΩ) must satisfy the following relationship (20)

G (jω) = j sinΩ. H (jω)

F (jω) = j sin Ω. G (jω) with Ω = T _A. ω _network (20) where ω is _{N etwork} the mains frequency of the monitored power transmission line and T _{A is} the sampling again. For small values of Ω it is approximately the case that sinΩ = Ω. This in turn means that the FIR filters of different types used with each other via the d / dt operator or in

Frequency range are linked via the p operator. The individual FIR filters can therefore be generated by folding a basic filter with a basic filter. The convolution theorem of the Fourier transform is used here. A transversal filter with a transfer function according to the following equation (21) is expediently used as the basic filter:

If a filter unit 16 in accordance with FIG. 5 is used to carry out an investigation using the FIR filters of the third type, analogously to the procedure according to equations (13) to (16), then a comparison length factor m _v can be analogous to equation (13) determine according to equation (22) below:

with a first additional size H5,

With

additional size H6, with

a third additional size H7 and a fourth additional size H8.

The index numbers "1" and "2" again identify the values of i _FA and I _OFA sampled at different sampling times.

Completely in accordance with what has already been explained above, there is a comparison error resistance R _V , which is shown by the following equation (23)

Accordingly, a comparative reactance X _V can be determined according to equation (24) below:

A comparison impedance is thus obtained. Since the acquisition of this comparison impedance takes place in parallel with the acquisition of the measurement impedance, when using the method according to the invention, after a measurement cycle for the distance measurement, one obtains a statement in the form of the measurement impedance and a further statement in the form of the comparison impedance. If both impedances are the same, then this is a clear statement that the Di punch measurement has been carried out precisely, so that a distance protection device working according to the method according to the invention can immediately decide on the basis of this distance measurement whether a triggering is to be carried out. In general, it is advisable to allow a certain deviation of the two measured impedance values; For example, it is considered justifiable to assume a sufficiently precise distance measurement in the event of a deviation of> 10%. A comparison is then made according to equation (25) below:

| XX _V | <0.1 | Z | (25)

If the comparison shows that the difference between the measurement impedance and the comparison impedance is relatively large, the measurement of measurement impedance and comparison impedance is repeated in a next measurement cycle; if the condition according to equation (25) is met, then a final protection decision is made.

Claims

claims

1.Method for performing a distance measurement on a multi-phase electrical power transmission line, in which - the voltage (U _RA ) on a faulty phase conductor is detected, digitized and in a linear-phase, non-recursive digital filter (FIR filter) (3) of a first type (with weighting factors g _i ) a filter unit (1) is evaluated,

- The current (i _FA ) in the faulty phase conductor is detected, digitized and, on the one hand, in a further FIR filter

(5) of the first type and on the other hand in an FIR filter

(6) a second type (with weight factors f _i ) of the filter unit (1) is evaluated,

- The weighting factors (g _i , f _i ) being freely specified

be and

- an error correction by means of a correction factor

(k _c ) is carried out, which is formed as a quotient from the amplitude responses of the FIR filters of the first and second type, and

- From the output variables of the filter unit (1) in one

Arithmetic unit (10) the respective distance of the fault location from a measuring point indicating impedance measured quantities are determined,

d a d u r c h g e k e n n z e i c h n e t that

- For distance measurement in the case of single-pole earth faults, a sum current (I _OFA ) corresponding to the sum of the currents in the phase conductors of the energy transmission line is recorded, digitized and on the one hand in an additional FIR filter (8) of the first type and on the other hand in an additional filter (9) of the second Type of the filter unit (1) is evaluated with formation of an output variable (m _k , n _k ),

- In the computing unit (10) from the output variable (w _k ) of the further FIR filter (5) of the first type, a first auxiliary variable (H1) by multiplication by the kilometric ohmic resistance (R ₁ ) of the co-system of the energy transmission line is formed,

a second auxiliary variable (H2) is formed in the computing unit from the output variable (V _k ) of the further FIR filter (6) of the second type by multiplication with the kilometric inductance (L ₁ ) of the co-system of the energy transmission line,

- in the arithmetic unit also a third auxiliary variable from the output variable (m _k ) of the additional FIR filter (8) of the first type by multiplication with the difference (R ₀ - R ₁ ) from the kilometric ohmic resistance of the zero system and the co-system of the energy transmission line ( H3) is obtained

- In addition, in the arithmetic unit, a fourth auxiliary variable (H4.) from the output variable (n _k ) of the additional FIR filter (9) of the second type by multiplying by the difference (L ₀ - L ₁ ) from the kilometric inductance of the zero system and the co-system of the energy transmission line ) is formed

in the computing unit from the output variables (y _k , m _k , n _k ,

w _k , v _k ) of the FIR filter (3,8,9,5,6) and the auxiliary variables (H1 to H4) a length factor (m) and a resistance value (Rf) proportional to the resistance at the fault location are calculated and

- in the arithmetic unit by multiplying the length factor (m) by the kilometric resistance

of the co-system and addition of the resistance value (R _f ) and by multiplying the kilometric reactance of the

Mitsystem with the length factor (m) is formed the measuring impedance (R, X) characterizing the distance of the fault location.

2nd Method according to claim 1,

d a d u r c h g e k e n n z e i c h n e t that

- In the presence of a parallel multi-phase electrical power transmission line one of the sum of the currents in the phase conductors of this energy transmission line, corresponding zero current (I _OAM ) is detected and digitized and

- A size proportional to the real part of the zero current at the input of the additional FIR filter (8) of the first type, the digitized total current (i _ok ) and one to the imaginary part of the zero current (l _OAM ) at the input of the additional FIR filter (9) of the second Type is added to the digitized total current (i _ok ).

3. The method according to claim 1 or 2.

characterized by the fact that

- Another distance measurement is carried out in parallel by

the voltage (u _RA ) on the faulty phase conductor is evaluated in a supplementary FIR filter (17) of a third type (with weighting factors (h _i ) in the filter unit (16) to form an output variable (o _k ),

- The current (i _FA ) in the faulty phase conductor in

a further complementary FIR filter (17) of the third

Type is evaluated with the formation of an initial auxiliary variable (p _k ),

- The total current (i _0FA ) is _evaluated in an additional supplementary FIR filter (19) of the third type with formation of an additional _auxiliary output variable (r _k ),

a first additional auxiliary variable (H5) is formed in the arithmetic unit (20) from the auxiliary output variable (p _k ) of the further supplementary FIR filter (18) of the third type by multiplication with the kilometric resistance (R ' ₁ ) of the co-system of the energy transmission line,

- In the computing unit (20) also from the output variable

(w _k ) of the further FIR filter (5) of the first type

Multiplication with the kilometric inductance (L ' ₁ ) of the co-system of the energy transmission line, a second additional variable (H6) is formed,

- in the computing unit (20) also from the auxiliary output variable (r _k ) of the additional additional FIR filter (19) of the first type, a third additional quantity (H7) is obtained by multiplying by the difference (R ₀ - R ₁ ) from the kilometric ohmic resistance of the zero system and the co-system of the energy transmission line,

- In the arithmetic unit (20) also from the output variable (r _k ) of the additional additional FIR filter (11) of the third type by multiplication by the difference (L ₀ - L ₁ ) from the kilometric inductance of the zero system and the co-system of the energy transmission line a fourth additional variable (H8) is formed,

- In the computing unit (20) from the output variables

(o _k , p _k , r _k ) of the FIR filter (17, 18, 19) and the additional variables (H5 to H8) a length factor (m _v ) and a resistance value (R _fv ) proportional to the resistance at the fault _{location is} calculated ,

- In the arithmetic unit (20) by multiplying the length factor (m _v ) by the kilometric reactance (L ' ₁ ) of the co-system with the length factor (m _v ) a comparison impedance (R _v , X _v ) characterizing the distance of the fault location is formed will, and

- The distance measurement is considered to be sufficiently accurate if the difference between the measurement impedance (R, X) and the comparison impedance (R _v , X _v ) is within a predetermined size.