WO1996032653A1 - Method of measuring distances along a high-tension power-transmission line - Google Patents

Abstract

The invention concerns a method for the measurement of distances along a high-tension power-transmission line, the method calling for measurements of the current and voltage in the power-transmission line to be digitalized and then processed in a digital filter unit consisting of two linear-phase, non-recursive FIR filters of a first type and at least one FIR filter of a second type. The individual weighting factors are predetermined independently, and an error correction carried out by applying a correction factor. From the processed output values from the filter unit, a computing unit determines parameters from which an impedance indicating the distance to the location of a fault can be deduced. In order to be able to obtain accurate distance measurements in the shortest possible time, the invention calls for the digitized measured values (u, x) in the filter unit (1) to be processed in two further FIR filters (8, 9) of a third type to determine, from the output signals from these FIR filters (8, 9), a reference impedance (Zv). The distance measurement is considered to be sufficiently accurate if the difference betweem the measured impedance (Zm) and the reference impedance (Zv) is less than a given value.

Description

description

Method for performing distance measurements on an electrical high-voltage transmission line

The invention relates to a method for performing distance measurements on an electrical high-voltage transmission line, in which measured variables derived from the current and voltage of the high-voltage transmission line are digitized and evaluated in a digital filter unit, which is made up of two linear-phase, non-recursive digital filters (FIR-Fil ¬ ter) of a first type with an amplitude response H (jω) and weight factors h ^ and consists of at least one FIR filter of a second type with an amplitude response G (jω) and weight factors g ^, the individual weight factors hj_, g ^ der FIR filters are freely specified, the error correction is carried out by means of a correction factor which is formed as a quotient from the amplitude responses H (Ω), G (Ω) of the FIR filters of the first and second type, which in the operating behavior of the High-voltage transmission lines describing differential equations occurring parameters from the evaluated outputs variables are determined in a computing unit and a measuring impedance indicating the respective distance to a fault location is determined therefrom and errors which have arisen during the evaluation are taken into account by an error correction.

Such a method is known from European patent EP 0 284 546 B1. In this method, the use of digital filters achieves a comparatively high accuracy when measuring the distance or the impedance to a point of failure on the high-voltage transmission line when implemented in a distance protection device. By choosing correspondingly fast computers, relatively short computing times can be achieved. However, FIR filters with a counter degree that is not too high should be selected, because otherwise the evaluation time or the time until a trigger signal is output to a switch in the high-voltage line becomes too long. This is somewhat at the expense of measuring accuracy, so that repeated measurement must be used in order to avoid false triggers. However, the use of the FIR filter guarantees that the required filter properties are present. The use of such filters also enables extremely simple error correction, which is also easy to integrate into the process flow.

The invention has for its object to provide a method for performing distance measurements on a high-voltage transmission line with FIR filters, with which accurate distance measurements can be carried out in an optimally short time.

To achieve this object, according to the invention, the digitized measured variables in the filter unit are evaluated in two further FIR filters of a third type with an amplitude response F (jω) and weighting factors fi in a method of the type specified at the outset, the filters of the various types NEN types the conditions G (jω) = j. sinΩ. H (jω) F (jω) = j. sinΩ. G (jω) with Ω = T_ ^. © _Net are sufficient, in which T___ denote the sampling time and ( _net the network frequency of the high-voltage transmission line; a comparison impedance is determined with the output _{signals of} the further FIR filters and the distance measurement is considered to be sufficiently accurate if the difference between the measurement impedance and holds the comparison impedance within a predefined size. The predefined size is to be understood as a fraction of the currently determined measurement impedance, so that - regardless of the measurement impedance measured in each case - the permitted relative difference between measurement and comparison impedance is always the same size. A major advantage of the method according to the invention is that it may be possible to dispense with repeated measurements if the first measurement already detects a difference between the measurement and comparison impedance that lies within a predetermined size. In a distance protection method, it is normally not possible to generate trigger commands for a first impedance value lying in a predetermined tripping polygon, because this is considered to be too unsafe. According to the invention, this uncertainty is eliminated by determining the comparison impedance in parallel over time and by forming the difference between the measurement and comparison impedance and monitoring this difference.

According to one embodiment of the method according to the invention, it is advantageous if FIR filters of the first type which can be described by a symmetrical counter polynomial (weight factor distribution h ^ = h _n _i) with an even counter degree _n and when using an antisymmetric counter polynomial (weight factor distribution g ^ = g _n -i) with ^a straight count degree n writable FIR filters of the second type, such filters are used as further FIR filters, such filters that can be described by a symmetrical counter polynomial (weight factor distribution h ^ = h _n _ι.). The numerator degree n can be chosen as an even or odd number; if it is chosen as an even number, the phase rotation caused by the filter, for which the degree n / 2 is the decisive factor, is multiplied by an integer.

Figures 1 and 2, of which serve to further explain the method according to the invention

FIG. 1A shows an equivalent circuit diagram of a high-voltage transmission line to be monitored as part of a distance measurement, 1B shows the representation of the normalized differential equation describing the behavior of the high-voltage transmission line in a block diagram, and FIG. 2 shows the principle of the method according to the invention in a block diagram.

The diagram of an electrical high-voltage transmission line shown in FIG. 1A has a network inductance L _f j and an ohmic network resistance R ^. It is traversed by a mains current ijj; A mains voltage u ^ can be tapped at the terminals of the high-voltage network. The quantities ijj and uj represent the output quantities of the high-voltage transmission line. The differential equation describing the high-voltage transmission line is:

_r di, _-.

UN = LN - + RN IN (i: German

This equation (1) can - as described in detail in the above-mentioned EP 0 284 546 - be converted into a representation independent of absolute values:

a x = -bx + u (2)

The block diagram for the standardized differential equation (2) can be found in FIG. 1B. The normalized parameters a and b, which have now taken the place of the absolute parameters Ljj and R ^, could be derived from equation (2)

Measurement of the quantities u, x and x can be determined at two different times t ^ and t2. The prerequisite for this is that the arrangement at times ti and t2 is in two mutually independent states.

As can also be gathered from EP 0 284 546 B1, the parameters a and b are determined without forming the first time derivative of the standardized current x by the standardized ones Output variables u and x are evaluated in a filter unit. Such an evaluation is carried out by convolution operations (symbolically represented by * in the block diagram according to FIG. 2). For this purpose, the normalized output variable u is converted to a digital FIR filter 3 via an analog-digital converter 2 (cf. FIG. 2), which converts the output variable u into a number sequence u ^ after sampling with a correspondingly selected sampling time T_ ^ fed. This digital FIR filter 3 belongs to a first filter type and has a symmetrical weight factor distribution h ^ = h _n _i. At the output of the digital FIR filter 3, a sequence yjς is created, the imaging rule of which is:

k-i (3) ι = 0

Furthermore, after corresponding scanning in a further analog-digital converter 4, the standardized output variable x is converted and the resulting values x _{jς are fed to} a digital FIR filter 5, which likewise belongs to the first filter type and whose weight factor distribution is identical to that of the digital one FIR filter 3; at its output a sequence wj _{ζ is} generated, which is described with: n

Wk = Σ hi 'Xk-i < ^{3 a} > i = 0

In addition, the sampled values xjς are fed to a digital FIR filter 6, which belongs to a second type, whose weight factor distribution is gi = g _n -i *.

At the output of this digital FIR filter 6, a sequence Vfc is created, the mapping rule of which is:

nv * = ∑ gi 'ki ⁽ 3 ^fe) i = 0 The output sequences ^v k » ^v k ^{and w} k generated by the FIR filters 3, 5 and 6 are linked to one another in accordance with the teaching of EP 0284 546 B1 such that they correspond to the differential equation (1) derived from the standardized differential equation (1). 4) are enough:

Equation (4) is a linear equation free of derivatives of the output quantity x with two unknowns, namely the parameters a and b, which is obtained by determining, evaluating and inserting two linearly independent samples of u and x is solvable and from which a measuring impedance Z _m can be determined. In particular, reference is again made to EP 0 284 546 B1.

According to the invention, the sequence of numbers ujς is evaluated by a folding operation in a further FIR filter 8 of a third type with weighting factors f ^, at whose output values m ^ arise which can be described by the following equation: n ntk = Σ /, ^•• Uk-i ⁽ 5 ⁾ ι = 0

Furthermore, the number sequence xjς is evaluated by means of a further FIR filter 9 of the third type; the following output variable n ^ results:

If the values _jς , n _jς and v ^ are then processed further in a _{computing module} 10 to determine a comparison _impedance Z _{v in} accordance with the procedure for determining the measurement impedance Z _m according to EP 0 284 546 _B1 , then the comparison impedance Z _v results in parallel to the measurement impedance. The variables corresponding to these two impedances Z _m and Z _v at the output of the arithmetic module 10 are fed to a difference former 11, which is followed by a limit value stage 12. Applies e.g.

I Zm - Z _v | <0.1. Z _m , (7)

the measuring impedance is then considered to be sufficiently precisely determined and a trigger signal is generated immediately without waiting for another measured value, that is to say without repeating the measurement.

Claims

claims 1. A method for performing distance measurements on an electrical high-voltage transmission line, in which - from current and voltage High-voltage transmission line derived measured quantities are digitized and evaluated in a digital filter unit (1), which consists of two linear-phase, non-recursive digital filters (FIR filters) of a first type (3.5) (with an amplitude response H (jω) and Weight factors h ^) and consists of at least one FIR filter of a second type (6) (with an amplitude response G (jω) and weight factors g ^), - The individual weight factors (h ^, gi) of the FIR filter (3,5,6) are freely specified, - The error correction is carried out by means of a correction factor (k _c ), which is formed as a quotient from the amplitude responses (H (Ω), G (Ω)) of the FIR filters of the first and second type (3,5,6), - the in the operational behavior of the High-voltage transmission lines descriptive differential equations occurring parameters are determined from the evaluated output variables in a computing unit and a measuring impedance indicating the respective distance to a fault location is determined and - Errors that have arisen during the assessment are taken into account by an error correction, that is, that - The digitized measured variables (u, x) in the filter unit (1) in two further FIR filters (8,9) of a third type (with an amplitude response F (jω) and weight factors fi) are evaluated, the filters of the different types meeting the conditions G (jω) = j. sinΩ. H (jω) F (jω) = j. sinΩ. G (jω) with Ω =

the sampling time and tön _et - denote the network frequency of the high-voltage transmission line, with the output signals

the further FIR filter (8, 9) determines a comparison impedance (Z _v ) and the distance measurement is considered to be sufficiently accurate if the difference between the measurement impedance (Z _m ) and the comparison impedance (Z _v ) remains within a predetermined size . 2. The method of claim 1, d a d u r c h g e k e n n z e i c h n e t that - at by a symmetrical counter polynomial (Weight factor distribution h ^ = h _n _i) with an even counter degree n writable FIR filters (3,5) of the first type and with an antisymmetric counter polynomial (weight factor distribution g ^ = g _n -i) with even counter degree n writable ( 6) of the second type, such filters are used as further FIR filters (8, 9), which are distinguished by a symmetrical counter polynomial (Weight factor distribution f ^ = f _n -i),