METHOD AND DEVICE FOR MONITORING A THREE-PHASE NETWORK OPERATED IN A COMPENSATED MANNER FOR A TUNING CHANGE OF THE ARC SUPPRESSION COIL

Abstract

A method for monitoring a three-phase network that is operated in a compensated manner for a tuning change of the arc suppression coil. Reference network parameters and a reference network frequency are determined for a tuned state, and a current network frequency is determined for a current state. A reference characteristic variable, which is proportional to a displacement voltage, is determined for the current network frequency using the reference network parameters, and a current characteristic variable, which is proportional to a displacement voltage, is determined at the current network frequency. A differential variable is determined from the reference characteristic variable and the current characteristic variable, from which, with a predetermined threshold value being exceeded, a tuning change is identified and changed network parameters are determined.

Claims

1-6. (canceled)

7. A method for monitoring a three-phase power network operated in a compensated manner for a tuning change of an arc suppression coil, the method which comprises: ascertaining reference network parameters and a reference network frequency for a tuned state; ascertaining a current network frequency for a current state; determining a reference characteristic variable, which is proportional to a displacement voltage, using the reference network parameters for the current network frequency; determining a current characteristic variable, which is proportional to a displacement voltage, at the current network frequency; determining a differential variable from the reference characteristic variable and the current characteristic variable, and identifying a tuning change from the differential variable when a predetermined threshold value is exceeded, and ascertaining modified network parameters.

8. The method according to claim 7, wherein each of the reference characteristic variable and the current characteristic variable is a displacement voltage.

9. The method according to claim 7, which comprises using a magnitude of a vectorial difference as the differential variable.

10. The method according to claim 7, which comprises time-delaying a determination of the modified network parameters if the differential variable continuously exceeds the predetermined threshold value over a predetermined period of time.

11. The method according to claim 7, wherein the network parameters comprise at least one inductance of the arc suppression coil and at least one line capacitance.

12. A device for monitoring a three-phase network for a tuning change of an arc suppression coil, the device comprising: a measuring device for ascertaining a displacement voltage and a network frequency; a computing device connected to said measuring device for determining network parameters; and wherein the device is configured to carry out the method according to claim 7.

Description

[0028] The invention is explained hereafter in more detail by means of an exemplary embodiment shown in the enclosed drawings. In the drawings:

[0029] FIG. 1 shows a simplified null equivalent circuit diagram of a three-phase system,

[0030] FIG. 2 shows examples of locus curves of displacement voltages and triggering thresholds when monitoring a neutral point voltage,

[0031] FIGS. 3-4 show examples of the frequency response of a three-phase network over time,

[0032] FIGS. 5-7 show examples of locus curves of a three-phase network.

[0033] FIG. 1 shows an example of a simplified null equivalent circuit diagram of a three-phase network.

[0034] A network asymmetry is illustrated, which is formed mainly by different conductor-earth capacitances in the three phases. In the null equivalent circuit, the unbalance current I.sub.u representing this unbalance is represented by the driving unbalance voltage E.sub.u and the unbalance impedance Z.sub.u.

[0035] To compensate for the capacitive fault current I.sub.f due to the ground fault, the neutral point of the network is earthed by means of an arc suppression coil L.sub.ASC (Petersen coil).

[0036] The network impedance Z.sub.0 shown in the equivalent circuit diagram is formed by the parallel connection of an ohmic equivalent resistance R.sub.0 for the entire network losses, including the losses of the arc suppression coil, the network capacitance C (line capacitance), and the inductance L.sub.ASC of the arc suppression coil.

[0037] The equivalent resistance R.sub.0 determines a damping current I.sub.D. The impedance of the arc suppression coil L.sub.ASC and thus the level of an inductive compensation current can be modified directly by adjusting the air gap in the iron core of the arc suppression coil, or the secondary side of the arc suppression coil L.sub.ASC is wired to an inductance, a capacitor or a defined current infeed, and/or the neutral point or one of the three phases can be wired to an ohmic resistor, an inductor, a capacitor or a defined current infeed. Such measures and/or devices are known to the person skilled in the art, so they do not need to be explained or described in more detail.

[0038] The coil position can be determined by measuring the position using a potentiometer. This results in a coil current I.sub.pos.

[0039] The tuning of the arc suppression coil L.sub.ASC takes place during the normal operation of the network (fault-free network condition), wherein the arc suppression coil L.sub.ASC is adjusted such that the inductive current through the arc suppression coil L.sub.ASC is the same as the capacitive current through the line capacitance (resonance current I.sub.res).

[0040] In practice, for an arc-suppressed network, but one without active residual current compensation in normal operation, a slight over-compensation or under-compensation can be set. This can be achieved by setting a current I.sub.pos which depends on the position of the plunger of the arc suppression coil L.sub.ASC.

[0041] With active residual current compensation, it may be appropriate to tune exactly to a resonance (I.sub.pos=I.sub.res).

[0042] With an exact tuning, the network impedance Z.sub.0=R.sub.0, which means that a maximum of the impedance of the parallel resonant circuit (L.sub.ASC, R.sub.0, C) is present, and the current through the fault location becomes minimal without residual current compensation.

[0043] Even with exact tuning, a complete compensation of the fault current is not possible with the arc suppression coil alone, since ohmic losses cannot be compensated by the arrangement.

[0044] These losses can occur by means of active residual current compensation, i.e. active current infeed with a compensation current I.sub.eci(t).

[0045] A complex displacement voltage U.sub.0 occurs at the arc suppression impedance Z.sub.0 and a displacement current I.sub.0 flows through the network impedance Z.sub.0.

[0046] The unbalance current I.sub.u, the current through the network impedance Z.sub.0 and the compensation or displacement current I.sub.edi converge at the neutral point of the three-phase network.

[0047] FIG. 2 shows examples of a locus curve N3 of the displacement voltage U.sub.0 and triggering thresholds N1, N2 when monitoring a neutral point voltage.

[0048] Threshold N1 represents a triggering criterion based on a change in the absolute value of the displacement voltage U.sub.0, and N2 indicates a triggering criterion based on the vectorial change in the displacement voltage U.sub.0 for a re-compensation of the network according to the prior art.

[0049] FIG. 3 shows a first example of the frequency response of a three-phase network over time.

[0050] It is a snapshot of the generally very stable European UCTE network, in which the frequency change is about 30 mHz within a period of five minutes.

[0051] FIG. 4 shows a second example of the frequency response of a three-phase network over time.

[0052] The curve shown is a snapshot of an Australian network, in which the frequency change is approximately 200 mHz within a period of five minutes.

[0053] Compared to FIG. 3, it can be seen that the Australian network fluctuates significantly more in frequency.

[0054] FIG. 5 shows a first example of a locus curve of a three-phase network in a 200.0 A-network with a circle drawn in for the trigger criterion.

[0055] At a damping current I.sub.D of 5 A, a frequency change of approximately 130 mHz can cause a vectorial voltage change of more than 20% and therefore, for example, the triggering criterion in the form of the corresponding circle radius in the figure will be met.

[0056] Calculation of the network admittance Y.sub.0 as a function of frequency:

[00001]$Y_o=\frac{1}{Z_o}$$R_o=\frac{V_{\mathrm{nom}}}{I_d}$$Y_{o\left(f\right)}=\underset{\mathrm{Losses}}{\underbrace{\frac{1}{R_o}}}+j.Math.\underset{\mathrm{Detuning}}{\underbrace{\left(w_{\left(f\right)}.Math.C-\frac{1}{w_{\left(f\right)}{L}_{\mathrm{ASC}}}\right)}}$$U_{\mathrm{nom}}.Math.Y_{o\left(f\right)}=U_{\mathrm{nom}}.Math.\frac{1}{R_o}+j.Math.\left(U_{\mathrm{nom}}.Math.w_{\left(f\right)}.Math.C-\frac{U_{\mathrm{nom}}}{w_{\left(f\right)}{L}_{\mathrm{ASC}}}\right)$$U_{\mathrm{nom}}.Math.Y_{o\left(f\right)}=I_d+j.Math.\left(U_{\mathrm{nom}}.Math.\frac{w_{\left(f\right)}}{w_{\left(f50\right)}}.Math.w_{\left(f50\right)}.Math.C-\frac{U_{\mathrm{nom}}}{\frac{w_{\left(f\right)}}{w_{\left(f50\right)}}.Math.w_{\left(f50\right)}.Math.L_{\mathrm{ASC}}}\right)$$k_f=\frac{w_{\left(f\right)}}{w_{\left(f50\right)}}$$Y_{o_{\left(f\right)}}=\frac{1}{U_{\mathrm{nom}}}.Math.\left(I_d+j.Math.\left(\underset{I_{v\left(f\right)}}{\underbrace{k_f.Math.I_{\mathrm{res}\left(f50\right)}-\frac{1}{k_f}.Math.I_{\mathrm{pos}\left(f50\right)}}}\right)\right))$

[0057] where a nominal voltage U.sub.nom is the phase voltage.

[0058] An angular frequency ω.sub.(f) refers to the current network frequency, which can exhibit a deviation from the nominal network frequency of 50 Hz, while an angular frequency ω.sub.(f50) refers to the nominal network frequency of 50 Hz.

W.sub.(f)=2.Math.π.Math.f

[0059] A weighting factor k.sub.f expresses the current frequency f as a proportion of the nominal frequency f50.

[0060] The relevant frequency change corresponds to the difference between the current frequency and the frequency during the tuning procedure.

[0061] In the above relationships, the network frequency during a tuning procedure was assumed to be 50 Hz, which is expressed by the index “f50”.

[0062] The considerations are also valid for networks with a different nominal frequency, such as 60 Hz.

[0063] After calculating the network parameters for each tuning procedure, including the unbalance impedance Z.sub.u, the parameters of the parallel resonant circuit Z.sub.0, formed from the inductance of the ground-fault arc suppression coil L.sub.ASC, the line capacitances C and the total network losses R.sub.0, are known more or less exactly.

[0064] Additional external inductors, such as fixed coils or distributed arc suppression coils, can be detected by the well-known multi-frequency method.

[0065] It is also possible to take a parametric approach to additional inductances, such as a fixed coil with a value at a nominal frequency and a switching state.

[0066] When calculating the network parameters via a coil adjustment or 50 Hz current infeed, only the detuning current I.sub.v can be determined; the back calculation to the network variable at the nominal or resonance frequency I.sub.res is performed via the measured coil position with I.sub.res=I.sub.pos−I.sub.v.

[0067] The network frequency is measured repeatedly, for example by measuring the reference voltage between two phases. The network parameters are expressed in terms of the measured frequency.

[0068] After the tuning procedure, the displacement voltage U.sub.0,ref is saved and used as a reference variable for the vectorial difference formation at the current frequency.

[0069] This measured value can be acquired with or without additional filters, or else parametrically based on the determined network parameters.

[0070] The reference displacement voltage U.sub.0,ref can now also be related to the currently measured network frequency via the network parameters in order to achieve frequency-dependent triggering and compensation.

[0071] Based on the relationships described above, the method according to the invention for monitoring a three-phase network operated in a compensated manner for a change in the tuning of the arc suppression coil can be summarized by the following method steps: [0072] reference network parameters and a reference network frequency are determined for a tuned state, [0073] a current network frequency is determined for a current state, [0074] a reference displacement voltage is determined from the reference network parameters for the current network frequency, [0075] a current displacement voltage is determined at the current network frequency, [0076] a differential variable is determined from the reference displacement voltage and the current displacement voltage using the magnitude of a vectorial difference, from which, when a predetermined threshold value is exceeded, a tuning change is identified and modified network parameters are determined.

[0077] The above steps can be executed repeatedly, wherein the determination of the modified network parameters is time-delayed if the differential variable continuously exceeds the predetermined limit over a predetermined period of time.

[0078] The calculation of the reference characteristic at the current frequency f can be carried out, for example, from the network parameters of the simplified null equivalent circuit according to FIG. 1.

[0079] The network impedance Z.sub.0(f) and the unbalance impedance Z.sub.u(f) can

be frequency-dependent.

[00002]$Z_{o\left(f\right)}=\frac{1}{Y_{o\left(f\right)}}$$U_{o\left(f\right)\mathrm{REFERENCE}}=\left(\frac{E_u}{Z_{o\left(f\right)}+Z_{u\left(f\right)}}+I_{\mathrm{eci}}.Math.\frac{Z_{u\left(f\right)}}{Z_{o\left(f\right)}+Z_{u\left(f\right)}}\right).Math.Z_{o\left(f\right)}$

[0080] The reference characteristic variable identified can therefore be used to form a difference value with respect to the measured displacement voltage.

[0081] Optionally, the differential variable can additionally be filtered, for example by means of averaging, or linear or even non-linear filtering.

[0082] FIG. 6 shows a second example of locus curves of a three-phase network in a 100.0 A-, 200.0 A- and a 300.0 A-network, with a circle drawn in for the trigger criterion for the latter network.

[0083] At a damping current I.sub.D of 5 A, a frequency change of approx. 90 mHz can cause a vectorial voltage to change by more than 20% and therefore, for example, the triggering criterion in the form of the corresponding circle radius in the FIG. will be met.

[0084] FIG. 7 shows a third example of a locus curve of a three-phase network in a 200.0 A-network.

[0085] Two circles are shown, one with its center at 50.0 Hz and one with its center at 50.2 Hz, which allow a frequency-dependent adjustment of the trigger circle.

[0086] In other words, the trigger criterion for a fresh determination of the network parameters is defined frequency-dependently.

[0087] As long as the triggering criterion is not exceeded by the current displacement voltage, the inductance of the arc suppression coil L.sub.ASC is kept constant.

LIST OF REFERENCE SIGNS

[0088] C network capacitance [0089] E.sub.u driving unbalance voltage [0090] I.sub.d damping current [0091] I.sub.eci compensation current (enhanced current injection, ECI) [0092] I.sub.pos current with respect to coil core position [0093] I.sub.res resonance current [0094] I.sub.u unbalance current [0095] I.sub.v detuning current (Iv=Ipos−Ires) [0096] L.sub.ASC inductance of the arc suppression coil (ASC) [0097] N1, N2 threshold value [0098] N3 locus curve of the displacement voltage U.sub.0 [0099] NW network [0100] R.sub.0 equivalent resistance [0101] U.sub.0, U.sub.0,res displacement voltage, zero voltage [0102] Z.sub.0 impedance of arc suppression coil [0103] Z.sub.u impedance unbalance