Differential protection method, differential protection device, and differential protection system

Abstract

A differential protection method for monitoring a line of a power grid. Current phasor measured values are captured at the ends of the line and transmitted to an evaluation device which is used to form a differential current value with current phasor measured values temporally allocated to one another. Time delay information indicating the time delay between local timers of the measuring devices is used for the temporal allocation of the current phasor measured values captured at different ends, and a fault signal indicating a fault affecting the line is generated if the differential current value exceeds a predefined threshold value. The reliability of the time synchronization is further increased by forming a quotient of the current phasor measured values to form an asymmetry variable, that is used to check a transit time difference of messages transmitted via the communication connection in different directions.

Claims

1. A differential protection method for monitoring a line of a power grid, the method comprising: capturing current phasor measured values with measuring devices at respective ends of the line, the measured values indicating an amplitude and a phase angle of a phase current flowing at a respective end of the line, wherein the measuring devices include local timers and allocate a timestamp to the current phasor measured values indicating a time of a capture of the current phasor measure values; transmitting at least the current phasor measured values captured at one end via a communication connection to an evaluation device; forming a differential current value through vectorial addition by the evaluation device with current phasor measured values temporally allocated to one another; determining time delay information indicating a time delay between the local timers of the measuring devices with a transit-time-based time synchronization method and using the time delay information for temporally allocating the current phasor measured values captured at different ends of the line; forming a quotient of the current phasor measured values temporally allocated to one another to form an asymmetry variable, and using the asymmetry variable to check a transit time difference of messages transmitted via the communication connection, the transit time difference arising on account of transmission times that are different depending on the direction; and generating a fault signal indicating a fault affecting the line if the differential current value exceeds a predefined threshold value.

2. The differential protection method according to claim 1, which comprises forming the quotient by dividing the current phasor measured value with a smaller absolute value by a current phasor measured value with a larger absolute value.

3. The differential protection method according to claim 2, which comprises forming the asymmetry variable only when the quotient of the current phasor measured values exceeds a predefined minimum value.

4. The differential protection method according to claim 2, which comprises: analyzing a position of the asymmetry variable in a complex number plane; and inferring a presence of a transit time difference based on the position.

5. The differential protection method according to claim 4, which comprises generating a fault message on a basis of a distance between the position of the asymmetry variable and a reference operating point.

6. The differential protection method according to claim 5, which comprises blocking the generation of the fault message indicating a fault affecting the line when the distance between the position of the asymmetry variable and the reference operating point exceeds a maximum distance.

7. The differential protection method according to claim 5, which comprises determining the position of the reference operating point based on a capacitive charging current of the line.

8. The differential protection method according to claim 4, which comprises: defining a region, within which the position of the symmetry variable is allowed, in the complex number plane; and suspending the checking of the transit time difference if the asymmetry variable lies outside the defined region.

9. The differential protection method according to claim 1, which comprises forming the asymmetry variable only when the absolute value of at least one of the current phasor measured values exceeds a capacitive charging current of the line by a predefined factor.

10. The differential protection method according to claim 1, which comprises forming the asymmetry variable only when the absolute values of the current phasor measured values are below a threshold dependent on a nominal current of the power grid.

11. A differential protection device for monitoring a line of a power grid, the device comprising: a measuring device having a local timer and being configured to capture current phasor measured values at one end of the line, the values indicating an amplitude and a phase angle of a phase current flowing at the one end of the line, and to allocate a timestamp to the current phasor measured values indicating a time of a capture thereof; a communication device configured to exchange current phasor measured values via a communication connection with another differential protection device; and an evaluation device configured to form a differential current value with its own current phasor measured values temporally allocated to one another on the one hand, and current phasor measured values received from the other differential protection device on the other hand, through vectorial addition, wherein time delay information indicating a time delay between the local timers of the measuring devices of the differential protection devices is used for temporally allocating the current phasor measured values captured at different ends, wherein the time delay information is determined using a transit-time-based time synchronization method, and to generate a fault signal indicating a fault affecting the line if the differential current value exceeds a predefined threshold value; wherein a quotient is formed of the current phasor measured values temporally allocated to one another, forming an asymmetry variable, and wherein the asymmetry variable is used to check a transit time difference of messages transmitted via the communication connection, the transit time difference arising on account of transmission times that are different depending on a direction.

12. A differential protection system for monitoring a line of a power grid, the system comprising: at least two differential protection devices each disposed at a respective end of the line; a communication connection connecting the individual said differential protection devices for a transmission of current phasor measured values; each of said at least two differential protection devices including: a measuring device having a local timer and being configured to capture current phasor measured values at a respective end of the line, the measured values indicating an amplitude and a phase angle of a phase current flowing at the respective end of the line, and to allocate a timestamp to the current measured values indicating the time of a capture thereof; a communication device configured to exchange current phasor measured values via said communication connection with a respective other differential protection device; at least one of said at least two differential protection devices including: an evaluation device configured to form a differential current value with its own current phasor measured values temporally allocated to one another and current phasor measured values received from the respectively other differential protection device, through vectorial addition, wherein time delay information indicating the time delay between the local timers of said measuring devices of said differential protection devices is used for the temporal allocation of the current phasor measured values captured at different ends, wherein the time delay information is determined using a transit-time-based time synchronization method, and to generate a fault signal indicating a fault affecting the line if the differential current value exceeds a predefined threshold value; and wherein said at least one of said at least two differential protection devices is configured to form a quotient of the current phasor measured values temporally allocated to one another to form an asymmetry variable, the asymmetry variable being used to check a transit time difference of messages transmitted via the communication connection, which transit time difference arises on account of transmission times which are different depending on a direction of the transmission.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

(1) FIG. 1 shows a schematic view of a differential protection system for monitoring a line of a power grid;

(2) FIGS. 2 and 3 show a schematic view of a differential protection system in the case of an external fault and in the case of an internal fault;

(3) FIG. 4 shows a schematic view of a differential protection system for explaining the transmission of current phasor measured values;

(4) FIG. 5 shows a graph for explaining the emergence of an angular error as a result of a transit time difference when transmitting the current phasor measured values;

(5) FIG. 6 shows a graph in the complex number plane for explaining the position of faulty and fault-free operating conditions;

(6) FIG. 7 shows a first graph for explaining the evaluation of an asymmetry variable;

(7) FIGS. 8 and 9 show circuit diagrams and graphs for explaining the influence of a capacitive charging current on the asymmetry variable;

(8) FIG. 10 shows a second graph for explaining the evaluation of the asymmetry variable, taking into account a capacitive charging current; and

(9) FIG. 11 shows a third graph having different regions for explaining the evaluation of the asymmetry variable.

DETAILED DESCRIPTION OF THE INVENTION

(10) Referring now to the figures of the drawing in detail and first, in particular, to FIG. 1 thereof, there is shown a part 10 of a power grid. The part 10 comprises a three-phase line 11 which may be designed, for example, as an overhead line or as a cable. The line 11 is monitored at its first end 11a by way of a first differential protection device 12a and at its second end 11b by way of a second differential protection device 12b for faults occurring on the line, such as, for example, short circuits. For this purpose, current signals are captured for each phase 13a, 13b, 13c of the line 11 with first current transformers 14a-c at a first measuring point at the first end 11a of the line 11 and second current transformers 15a-c at a second measuring point at the second end 11b of the primary component 11 and are fed to a respective measuring device of the differential protection devices 12a, 12b. Current phasor measured values which provide an indication of the amplitude and phase angle of the current signal at the time of capture are generated from the analog current signals. The current phasor measured values, including an A/D conversion, can be generated in the measuring device of the respective differential protection device 12a, 12b, in the current transformers themselves or in a suitable interposed measuring device (not illustrated in FIG. 1), e.g. a Phasor Measurement Unit (PMU), a Remote Terminal Unit (RTU) or a Merging Unit. Finally, the generated current phasor measured values are fed to an evaluation device, e.g. a CPU or a signal processor, of the respective differential protection device 12a, 12b.

(11) The differential protection devices 12a and 12b are interconnected by means of a communication connection 16 which is indicated only schematically in FIG. 1 and may be, for example, an IP-based communication network or a telecommunication network. However, any other communication connection of any type can also be used to connect the differential protection devices 12a and 12b. The respective differential protection device 12a or 12b can be supplied via this communication connection 16 with the current phasor measured values from the respective other end 11a, 11b of the line 11, i.e. pairs of current phasor measured values recorded at both ends 11a and 11b can be formed in each case in each differential protection device 12a and 12b for each phase 13a, 13b, 13c of the line 11.

(12) Using the current measured values from both ends 11a and 11b of the primary component 11 which are available in both differential protection devices 12a and 12b, a differential current value can be formed in one or both differential protection devices 12a and 12b by means of the evaluation device through vectorial addition of the current phasor measured values and subsequent absolute value formation for each phase, and can be compared with a threshold value.

(13) In the case of a fault-free line 11, the current entering the line 11 for each phase is (more or less) equal to the current flowing from the line 11, so that a phasor with the absolute value of around zero should be obtained through vectorial addition of the current phasor measured values. However, in reality, the differential current value will always assume a value that is not equal to zero, but will be below a predefined threshold value. This can be attributed, for example, to a capacitive charging current on the monitored line, which causes the differential current value to virtually never permanently assume exactly the value zero, even in the fault-free case. In addition, transformer inaccuracies and measurement errors, for example, can also contribute to this effect. The predefined threshold value can be specified as either static or dynamic, for example adapted to the level of the respective phase currents.

(14) The threshold value can be specified as a separate parameter. However, it can also be provided to check whether the threshold value has been exceeded by evaluating the position of a measured value pair consisting of the differential current value and an associated stabilization value in a tripping diagram. For this purpose, differential current values and associated stabilization values are formed from associated, i.e., simultaneously captured, current phasor measured values and the position of the measured value pair consisting of a differential current value and a stabilization value is checked in the tripping diagram. If the measured value pair is located within a tripping range, a fault affecting the monitored line is inferred and the fault signal is generated.

(15) If the differential current value exceeds the predefined threshold value for a specific phase, this indicates a fault affecting the relevant phase of the line 11, which may, for example, be a short circuit to ground or a two-pole or multi-pole short circuit, i.e., a short circuit between two or more phases of the primary component. For that phase in which the fault has been detected, the differential protection devices 12a and 12b generate a fault signal, as a result of which the emission of a tripping signal is effected via control lines 17a, 17b to phase-selectively switchable circuit breakers 18 and 19. The tripping signal causes the corresponding phase-related circuit breaker 18a, 18b, 18c or 19a, 19b, 19c to open its switching contacts, so that the phase 13a, 13b, 13c affected by the fault is disconnected from the remainder of the power grid.

(16) If, for example, a short circuit to ground occurs on the phase 13b, the differential protection devices 12a and 12b detect this on the basis of a differential current value exceeding the respective threshold value and transmit tripping signals to the phase-related circuit breakers 18b and 19b in order to disconnect the phase 13b of the line 11 from the power grid.

(17) Although a three-phase line 11 with only two ends 11a and 11b is shown in FIG. 1, the method according to the invention can also be used with any single-phase or multi-phase lines with two or more ends, for example electrical busbars with a plurality of branches.

(18) Furthermore, notwithstanding the illustration according to FIG. 1, it can also be provided that the current phasor measured values are transmitted to a single differential protection device and are evaluated there. In this case, it suffices to place measuring devices at the ends 11a, 11b of the line 11 to capture the current phasor measured values and transmit them to the differential protection device. This differential protection device could be disposed at one of the line ends, but also at any other position, for example as a central differential protection device in a switchgear station or control station or else in a data processing cloud.

(19) In order to be able to determine the differential current value correctly, it is necessary for the current phasor measured values used for its formation to have actually been simultaneously captured at the ends 11a, 11b of the line 11. However, a time delay normally occurs, particularly when transmitting the current phasor measured values over a comparatively long communication route, so that the locally captured current phasor measured value cannot readily be linked to a current phasor measured value captured at a distant end and transmitted. If current phasor measured values that have not been simultaneously captured are used, differential current values may occur which exceed the threshold value and would therefore result in the emission of a fault signal, even in a line that is actually fault-free.

(20) For the temporal allocation of the current phasor measured values, the latter are therefore normally provided with a marking in the form of a timestamp which indicates the time of their capture. By selecting those current phasor measured values from different ends of the line which have a matching timestamp, it can be ensured that the differential current value is calculated correctly. However, a prerequisite for this procedure is that the measuring devices used to capture the current phasor measured values in each case have local clocks or timers (CLK) which are synchronized with one another or at least have a known time delay. In order to achieve this, any time delay between the timers (CLK) of the respective measuring devices is continuously determined and is either used to readjust a timer (CLK) or is used by the evaluation device of the differential protection device for the temporal allocation of the current phasor measured values. In the last-mentioned case, for the temporal allocation of the current phasor measured values, the determined time delay must be subtracted from the timestamp of the current phasor measured value of that measuring device which has the timer with the time that is ahead of the other timer (CLK).

(21) A transit-time-based synchronization method, for example the so-called “ping-pong method”, can be used to determine the time delay between the timers (CLK) of the measuring devices. That time duration which is required to transmit a first message in one direction and then a second message in the other direction via the transmission route between the two measuring devices is measured here. In each case, the transmitted messages have a timestamp which indicates the time of their dispatch. The measuring devices furthermore record the reception time of the respective message. The time duration for the pure transmission of the messages (without any time delays between the reception of the first message and the dispatch of the second message) can be determined by means of the timestamps. The determined time duration is halved and provides the transit time on the communication route for a message transmitted in one transmission direction. The measuring devices can determine the time delay between the timers (CLK) of the measuring devices by means of the timestamps transmitted with the messages and the reception times and the transit time which is now known. Further details of the ping-pong method can be found in the above-mentioned patent No. U.S. Pat. No. 8,154,836 B2.

(22) However, the determination of the time delay using a transit-time-based method can supply reliable results only if the communication route between the measuring devices is symmetrical, i.e. if the transit times of the messages for the forward path and the return path over the communication route are identical. In the case of an asymmetrical communication route, i.e. non-identical transit times for the forward path and the return path, the method supplies an incorrect transit time, so that the time delay determined using the transit time is also erroneous. In this case, current phasor measured values which are not simultaneously captured are erroneously used to calculate the differential current phasor value. In the worst case, this can result in the determination of a differential current phasor value which exceeds the threshold value despite an actually fault-free line.

(23) It must therefore be ensured that an immediate detection takes place if a communication route is asymmetrical from the outset or changes, gradually or abruptly, from a symmetrical communication route to an asymmetrical communication route. For example, a previously symmetrical communication route can assume an asymmetrical behavior due to switching operations of switches or routers which modify the communication path of a message. Ageing effects or topology changes can also affect the behavior of a communication route.

(24) According to the method proposed below, the occurrence of an asymmetrical behavior of a communication route or other circumstances distorting the transit time measurement can be comparatively easily detected. Only the current phasor measured values which are in any case present are required for this purpose.

(25) As already described, in line differential protection, the current phasor measured values on overhead lines and cables with two or more ends are measured in a phase-selective manner and the vectorial sum of the current phasor measured values is then formed across all measuring stations. According to Kirchhoff's node rule, this sum must always be zero. This applies to the fault-free state of the line and likewise in the case of an external fault. If a fault occurs in the protection region (internal fault), that is to say on the line itself, differential currents which are not equal to zero arise. The line differential protection determines a fault on the basis of this criterion and transmits a tripping signal to a circuit breaker which disconnects the fault.

(26) In this context, FIG. 2 shows the simplified single-phase basic circuit diagram of the differential protection for two ends of a line 11 for the case of an external fault, while FIG. 3 shows a corresponding basic circuit diagram for the case of an internal fault. The fault is indicated in each case by a lightning symbol 20. Only one differential protection device 12 is illustrated in a simplified manner in FIGS. 2 and 3. The indices L and R indicate the reference position. IL corresponds to the current phasor measured value at the local end of the line and IR corresponds to the current phasor measured value at the remote end of the line. It should be noted that currents which flow into the protection region are positively counted and currents flowing away are negatively counted. The short-circuit currents of the line are provided with the index K.

(27) However, influences which result in a differential current value different from zero occur even during fault-free operation. These may be, for example, saturation phenomena of the current transformers, capacitive charging currents or asynchronous communication of the line. For this reason, a response value of zero cannot be accepted.

(28) For long lines, the current phasor measured values must be transmitted by means of serial communication, for example by means of optical waveguides. This is schematically indicated in FIG. 4 in which the differential protection devices 12a and 12b transmit their current phasor measured values IL and IR to the respective other differential protection device 12a, 12b via the communication connection 16. Alternatively, the transmission can also take place to only one of the differential protection devices 12a, b or to a central device.

(29) The so-called “ping-pong” method, for example, is used to synchronize the phasor values, but this presupposes identical transmission times tS1 for the forward path and tS2 for the return path (tS1=tS2). If differences (asymmetries) occur in the transit times, undesirable differential currents arise. The current phasor measured values IL and IR are rotated by an erroneous angle ΔΘ, as can be seen in FIG. 5 (IRΔΘ is entered as a rotated current phasor), with the result that an apparent differential current IDiff not equal to zero is also produced in the fault-free case. The resulting erroneous angle ΔΘ and erroneous currents resulting therefrom are described above in connection with equations (1) to (3).

(30) The aim is therefore to detect this interfering influence and to thereby monitor the method of operation of the differential protection. For this purpose, the complex current phasor measurement variables of the line ends are represented in a ratio with respect to one another in order to form an asymmetry variable AS. It is preferred in this case for the current phasor measured value with the smaller absolute value to be divided by the current phasor measured value with the larger absolute value. In the case of a line having two line ends, such a quotient would therefore be formed as follows:

(31) $AS=\frac{{\underset{\_}{I}}_L}{{\underset{\_}{I}}_R}\mathrm{for}.Math.{\underset{\_}{I}}_R.Math.>.Math.{\underset{\_}{I}}_L.Math.\mathrm{and}AS=\frac{{\underset{\_}{I}}_R}{{\underset{\_}{I}}_L}\mathrm{for}.Math.{\underset{\_}{I}}_L.Math.>.Math.{\underset{\_}{I}}_R.Math.$

(32) Ideally, that is to say without internal faults and interfering influences, the current phasor measured value at one end corresponds to the negative current phasor measured value at the other end. The asymmetry variable in the complex notation would therefore be:

(33) $AS=\frac{{\underset{\_}{I}}_R}{{\underset{\_}{I}}_L}=\frac{{\underset{\_}{I}}_R}{{\underset{\_}{-I}}_R}=-1+0j$

(34) The result can be represented in the complex number plane in a unit circle 50, as indicated in FIG. 6. Since the current phasor measured value with the smaller absolute value is always divided by the current phasor measured value with the larger absolute value, the asymmetry variable produced in this case cannot leave the unit circle 50. The ideal position of the asymmetry variable in the fault-free case (only electrical load) or in the case of external faults is indicated by point 51. This ideal position is also referred to as a reference operating point in the case of negligible capacitive charging currents. In this case IR=−IL. The regions of internal faults with single-ended supply (region 52) and with double-ended supply (region 53) are marked separately in the unit circle.

(35) If asymmetries in the transit times of messages in different transmission directions of 2 and 4 ms occur in a 60 Hz grid, angular deviations of 21.6° and 43.2° arise starting from the reference operating point. The absolute values of the current phasor measured values do not change. In order to ensure symmetrical transit times, a monitoring function is therefore proposed, which monitoring function monitors the angular deviations or phase rotations in the case of constant absolute values of the current phasor measured values on the left-hand side of the unit circle. For example, the monitoring function can detect and accordingly report deviations of up to 6 ms. In order to ensure a tolerance with respect to absolute value deviations of the current phasor measured values, a range of, for example, 0.9<|AS|<1 can also be covered. An appropriately dimensioned monitoring region 61 is entered in a unit circle 60 in FIG. 7. The reference operating point of the asymmetry variable is indicated by 62. The monitoring of the transit times is active only within the monitoring region 61 (for example at points 63 and 64) and it is no longer possible to reliably distinguish between an actual fault and an excessive transit time difference outside said region (for example at point 65). Depending on the distance (in degrees) between the position of the asymmetry variable and the reference operating point, different levels of the urgency of a fault message can be generated.

(36) In reality, in particular in the case of long lines, capacitive charging currents likewise result in angular errors. In addition, they cause errors in the current absolute values. The method of operation of the monitoring function can be restricted thereby. Capacitive charging currents Ic on overhead lines or cables can be determined as follows:

(37) $\begin{array}{cc}{\underset{\_}{I}}_c=\frac{U_{LL}}{\sqrt{3}}.Math.2\pi .Math.f_n.Math.C_b^{\prime }.Math.l& \left(4\right)\end{array}$
where

(38) I.sub.c: charging current value to be determined;

(39) U.sub.LL: conductor-conductor voltage of the power grid;

(40) f.sub.N: nominal frequency of the power grid;

(41) C.sub.b′: capacitive coating of the line;

(42) l: length of the line.

(43) If the load current is low in comparison with the capacitive charging current, rotations with respect to the reference operating point may occur on the basis of the supply (single-ended, double-ended), as indicated, for example, in FIGS. 8 (double-ended supply) and 9 (single-ended supply). In this case, FIGS. 8 and 9 each show a basic circuit diagram of a differential protection system and, beside this, the effects of a shift of the reference operating point as a result of the capacitive charging current in the unit circle in the complex number plane.

(44) Specifically, the rotation indicated in FIG. 8 results from the relationship represented in equation (5) in the case of double-ended supply:

(45) $\begin{array}{cc}AS=\frac{{\underset{\_}{I}}_R}{{\underset{\_}{I}}_L}=a+bj=\frac{-I_{\mathrm{Load}}+\frac{{\underset{\_}{I}}_C}{2}}{I_{\mathrm{Load}}+\frac{{\underset{\_}{I}}_C}{2}}& \left(5\right)\end{array}$

(46) In the case of values of the capacitive charging current IC which are high in comparison with the load current Load, the value of equation (5) tends toward the value 1 and therefore the right-hand side of the unit circle. As shown in FIG. 6, the value of the asymmetry variable is also there, however, in the case of an actual fault, with the result that a clear statement cannot be made.

(47) In addition, the rotation indicated in FIG. 9 results from the relationship represented in equation (6) in the case of single-ended supply:

(48) $\begin{array}{cc}AS=\frac{{\underset{\_}{I}}_R}{{\underset{\_}{I}}_L}=a+bj=\frac{-I_{\mathrm{Load}}}{I_{\mathrm{Load}}+{\underset{\_}{I}}_C}& \left(6\right)\end{array}$

(49) In the case of values of the capacitive charging current IC which are high in comparison with the load current Load, the value of equation (6) tends toward the value 0 and therefore the center of the unit circle. As shown in FIG. 6, a clear statement is also not possible for this region of the unit circle.

(50) As the result of the consideration, it can therefore be stated that the influence of the capacitive charging current increases with a falling load current and has the potential to shift the reference operating points to the coordinate origin or to the right-hand half-plane depending on the supply. Since it could be gathered from the previous consideration of the fault states in conjunction with FIG. 6 that the transit times of messages can be monitored only in the region of the left-hand half-plane of the unit circle without influencing the function of the differential protection, it can be determined that the transit times can be monitored only within particular ranges of the capacitive charging current.

(51) For example, a minimum current threshold of the charging current in relation to the flowing load current should be complied with. If the absolute value of at least one of the current phasor measured values (either the local current phasor or the remote current phasor) is above a threshold value IS which results from the capacitive charging current multiplied by a factor x, the transit time monitoring is enabled:
Max(|I.sub.L|,|I.sub.R|)>I.sub.s, with I.sub.s=x.Math.|I.sub.C|

(52) A suitable value for x may be x=2.3, for example. This ensures that the influence of the charging current is low enough, such that the resulting reference operating point of the asymmetry variable still occurs in the left-hand half-plane of the unit circle in the fault-free case.

(53) If the current phasor measured values therefore meet the condition stated above, and at least one of the absolute values of the current phasors is therefore more than 2.3 times the absolute value of the charging current, the monitoring function is enabled.

(54) In the case of this value, the maximum possible load-dependent (resistive, inductive or capacitive) angular deviation would be approximately 25.4° starting from the reference operating point without a capacitive charging current (FIG. 7). This ensures that the position of the asymmetry variable is in the left-hand half-plane irrespective of the connected load.

(55) In order to make the method even more reliable, the ratio of the absolute values of the current phasor measured values in the fault-free case or in the case of external faults should also not be below a predefined minimum value, for example the value 0.56. This limit is illustrated, by way of example, as the left-hand edge of the fault region 52 in FIG. 6 and forms the limit value of the sensitivity of the differential protection in the case of internal single-pole faults with a high fault contact resistance.

(56) In the case of external faults which result in a high current flow, the influence of the capacitive charging current falls in contrast. The exact determination of the charging current fails. In these cases too, the monitoring function should be deactivated. It appears to be useful to limit the operating range of the monitoring function to an upper value, for example 1.5 times the nominal current:
|I.sub.L|≤1.5.Math.I.sub.N and |I.sub.R|≤1.5.Math.I.sub.N

(57) If load and line data are known, the level of the currents on the input and output sides can be calculated from the determined capacitive charging current calculated according to equation (4) from an appropriate grid model. It is then possible to determine the corresponding reference operating point therefrom under the influence of the capacitive charging current. The monitoring of the transit times takes into account the new reference operating point and accordingly adapts the monitoring region. In this respect, FIG. 10 shows a graph which corresponds to the illustration in FIG. 7 and has a monitoring region 91 which has been determined taking into account the capacitive charging current. Both the angular rotation and the absolute value of the reference operating point 92 have been shifted by the charging current in comparison with the reference operating point 62 in FIG. 7. The angular rotation of the reference operating point as a result of capacitive charging currents may be at most 25.4° in the example illustrated in FIG. 10. In addition, the tolerance of the absolute value differences for the asymmetry variable is extended to 0.8<|AS|<1. A tolerance of 10% in both directions is therefore still retained.

(58) Overall, a functionality for monitoring the transit times of messages used for time synchronization via the communication connection is implemented here in parallel with the differential protection function. Monitoring the position of the asymmetry variable (angular deviation in degrees) based on a reference operating point makes it possible to directly infer asymmetrical communication (the asymmetry in ms can be determined from the phasor rotation in degrees). The indication of asymmetry which occurs is that the position of the asymmetry variable moves away from the known reference operating point by a particular angle for the fault-free case, but is still in the indicated monitoring region. This means that the absolute values of the current phasor measured values have not changed with respect to one another. The monitoring region can be subdivided in order to react differently for different distances between the asymmetry variable and the reference operating point. This is indicated in FIG. 11. The known unit circle 101 is shown in FIG. 11. The reference operating point 102 has been calculated for the case of an existing capacitive charging current. A monitoring region 100 defines the permitted operating range of the transit time monitoring. The monitoring function is blocked outside the monitoring region 100, in the region 107, in order not to allow any overlaps with actual faults. Regions of different reaction to a deviation are depicted symmetrically around the reference operating point 102.

(59) A normal region 103 indicates permitted deviations which do not need to result in a reaction. In regions 104 of low communication asymmetries (for example at 250 μs-500 μs), it is possible to transmit a message containing an indication of the existing slight asymmetry. In the case of stronger asymmetries (region 105, at 0.5 ms-2 ms), it is possible to provide a fault message which comprises a warning, and, in the case of excessively high asymmetry or in the case of asymmetries of more than 2 ms (region 106), the differential protection can be completely blocked.

(60) If the monitoring region 100 is left, another fault is present. There is either actually an internal fault, to which the differential protection must respond, or there is a measurement error, for example in the case of current transformer saturation. In these cases, the monitoring function is deactivated.

(61) The differential protection can be set to be more sensitive by additionally monitoring the transit times since transit time differences which can affect the time synchronization can now be detected directly and in terms of their level. Transit time changes may be detected irrespective of whether they are abrupt or gradual. As a result of the fact that the current phasor measured values are represented in relation to one another directly in the complex plane, absolute value and angular errors can be considered separately from one another. In the case of an internal fault, the monitoring functionality is deactivated since the position of the asymmetry variable moves away from the monitoring region.

(62) Although the invention has been described and illustrated more specifically in detail above by means of preferred exemplary embodiments, the invention is not limited by the disclosed examples and other variations may be derived herefrom by a person skilled in the art without departing from the scope of protection of the patent claims set out below.