Title of Invention

"METHOD AND DEVICE FOR MONITORING AN ELECTRODE LINE OF A BIPOLAR HIGH-VOLTAGE DC TRANSMISSION SYSTEM"

Abstract

Method and device for monitoring an electrode line of a bipolar high-voltage DC transmission system In order to monitor an electrode line (4), comprising two lines (20, 22), of a bipolar HDT system, a balanced-to-earth pulse is fed from an unbalanced-to-earth pulse (u(t)) in push-pull mode into the lines (20, 22), and an actual echo curve (EK) which is compared with a dynamic desired echo curve is recorded from the echo signals, a fault signal being generated upon overshooting of a tolerance band placed around the echo differential curve (EDK). The result is a monitoring method which is almost independent of the strongly fluctuating earth conductivity, which has a longer range in simultaneous conjunction with a lower dispersion of the echo, and which radiates less electromagnetic energy and requires no additional decoupling elements.

Full Text

Description Method and device for monitoring an electrode line of a bipolar high-voltage DC transmission system The invention relates to a method and device for monitoring an electrode line of a bipolar high-voltage DC transmission system, this electrode line being split up from a branch point on into two lines. A system for transmitting power by means of a high-voltage direct current includes two converter stations which are interconnected by a DC line. In the case of so-called monopolar DC transmission, the two stations are interconnected by means of a single DC line, the reverse current is led through the earth. One DC pole in each station is then earthed by means of a good earth connection. Normally, this earth connection is arranged at a distance from the converter station and connected to the station by means of a line which is designated as electrode line. It can often be desirable or necessary to arrange the earth connection at a large distance of up to a hundred kilometres from the station. In the case of so-called bipolar DC transmission, the stations are interconnected by two DC lines, with the result that there is no need in normal operation for the direct current to be led through the earth. For various reasons, inter alia in order to permit monopolar operation of the system in the case of a converter failure, provision is also made of converter stations employing bipolar DC transmissions with an earth connection which is connected to the station by means of an electrode line. An electrode line is insulated with respect to the earth and normally comprises a line which is suspended on insulators. Although the voltage between the electrode line and earth is normally small in relation to other voltages in the system, an earth fault on the electrode line poses the risk of injury to - 2 - people or damage to other system components, for example corrosion damage. It is therefore necessary for earth faults, including high-resistance earth faults, and line breakages to be capable of being discovered quickly and reliably. It has already been proposed to use a differential protective arrangement to locate earth faults on an electrode line. In such a protective arrangement, the current is measured at both ends of the electrode line, and a difference between the two measured currents means that an earth fault is present. Such a protective arrangement has, however, various disadvantages. It requires a telecommunication connection between the two ends of the electrode line, and is therefore expensive, particularly in conjunction with long electrode lines. Such a protective arrangement also does not react to an earth tauit which occurs in cases in which the electrode line is conducting no current, something which is normally the case in undisturbed operation of bipolar transmission. In this case, as well, that is to say when no direct current flows through the electrode line, unbalanced currents can lead to the production of dangerous voltages on the line. It has further been proposed to locate earth faults on an electrode line by feeding an AC signal or an AC voltage signal of specific frequency into the line in the converter station. In this case, suppression filters are arranged at both ends of the line", these filters being tuned to the frequency of the introduced signal. An impedance measuring element serves to measure the impedance of the electrode line with respect to the earth at the feed point at the feed frequency. A change in the impedance thus measured is An indication of an earth fault. This method works well in the case of short electrode lines, but exhibits disadvantages for long electrode lines. In order to detect a line fault, the measuring frequency must be selected such that the length of the line is smaller - 3 - than a quarter of the wavelength. For this reason, it is necessary in the case of long electrode lines to select a frequency which is so low that there is the risk that the measurement will be disturbed by the system frequency or by the lowest harmonics of the system frequency. Furthermore, in the case of these low frequencies the suppression filters arranged at both ends of the electrode line, which must be dimensioned for the maximum current on the electrode line, become very large and expensive. EP 0 360 109 Bl specifies a protective device for an electrode line of the type mentioned at the beginning, in which a hiqh measuring frequency can be used even in the case of long electrode lines, as a result of which the dimensions and costs of the suppression filters and the risk of a disturbance by the system frequency or its harmonics are substantially reduced. In order to avoid standing waves on the electrode line, the suppression filter is provided at the remote end, referred to the feedpoint, of the electrode line with resistance elements which have a resistance value such that the filter is adapted to the characteristic impedance of the electrode line. This prevents the measuring signal from being reflected at the remote end of the electrode line. US-A 5,083,086 has disclosed a method for locating the point of a fault in a cable. In this method of determining the point of a fault, a repair specialist carries out this method, the defective cable firstly being isolated, that is to say the cable is not in operation. The next step is to connect at one end of the isolated cable a device bv means of which the method for determining the fault location is carried out. This device feeds a first electric pulse into the cable and records the reflections received, meieaicer, a voltage connected to the isolated cable is increased, a second pulse is fed into the cable and the received reflections are recorded. The increase in the feed voltage varies the impedance at the point of a fault in - 4 - the cable, with the result that it is possible to receive a reflection which uniquely reproduces the fault location. The recorded echo signals are intercompared. The fault location in the cable can then be calculated by means of this differential signal and a detected propagation time. An older national patent application with the official reference of 196 50 974.2 has disclosed a method and a device for detecting a state of an electrode line of a high-voltage DC transmission system (HDT system) . The HDT system is a bipolar HDT system. In this state-detecting method, a first electric pulse is fed in at a first end of the electrode line, and an echo signal of this line is detected. Thereafter, a second pulse is fed into the line at the. first end, and its echo signal is detected. These two echo signals are subsequently intercompared. An appropriate report signal is generated in the event of a deviation and/or agreement between the two echo signals. These method steps are continuously repeated until a fault signal is generated. The state-detecting method is stopped with this report signal. Recorded echo signals can then be used to determine the fault location. A comparison of the defective echo signal with stored echo signals for different operating situations permits the error (earth fault, line breakage, . . . ) to be determined more quickly. The device for detecting a state of an electrode line has a pulse generator, an evaluation device and a coupling element. The pulse of the pulse generator is fed into the electrode line via this coupling element, and its echo signal is relayed to the evaluation device. This device is connected to a first end of the electrode line. The second end of this electrode line is connected to earth potential. The electrode line is provided at the end with dampers in order for the electric pulse to run not into the HDT system, but only into the segment of the electrode line to be monitored. The evaluation device comprises a - 5 - comparator, a memory and a triggering device. Synchronously with a clock pulse, the pulse generator generates square-wave pulses which are affected by direct components and are fed continuously into the electrode line until a fault signal is present. This method permits simple fault detection during operation of the HDT system without the need to use existing measuring signals. This method therefore operates independently. Since in the case of no fault the earth participates on the line of the pulse, fluctuating earth conductivity exerts an influence on the echo signals, and thus on a reliable detection of falllts. Moreover, the radiation of electromagnetic energy, caused by the pulse in the common mode is very high. A further disadvantage consists in the fact that it is necessary to connect dampers in series into the electrode line at both ends of this electrode line. This makes for a very high outlay when retrofitting an existing HDT system. It is therefore the object of the invention to specify a method and a device for monitoring an electrode line of a bipolar HDT system, the disadvantages of the prior art which have been set forth no longer occurring, This object is achieved according, to the. , invention with the aid of the features of a method for monitoring an electrode line a bipolar high voltage DC transmission system and a device for monitoring the electrode line which will now be explained in detail, By virtue of the fact that a balanced-to-earth pulse is generated from an unbalanced-to-earth pulse in push-pull mode and is fed into the two lines of the electrode line, the earth scarcely participates any more on the line of these pulses, with the result that the method according to the invention is virtually independent of a strongly . fluctuating earth conductivity. A further advantage consists in the fact that the radiation in the form of electromagnetic energy is distinctly reduced by comparison with a common mode. Moreover, the push-pull mode causes a slight line attenuation, thus permitting a longer range - 6 - of the system in simultaneous conjunction with lower dispersion of the echo signal. The most important advantage of the push=pull mode is, however, its complete decoupling from the common mode. Interference signals which come from the HDT system can, however, propagate only in the common mode, since on this side of the branch point the electrode line is combined to form one conductor, with the result that an electromagnetic field can exist only between this conductor and earth. Interference signals coming from the HDT system propagate virtually at the speed of light on the electrode line, are split up at the branch point virtually equally in terms of amplitude and phase, and then migrate onto the two waveguides, specifically conductor-earth,and conductor-earth, to the end of the electrode line remote from the system. However, these interference signals cannot generate a voltage between the feed terminals fitted at equal spacings from the branch point, for which reason there is an ideal decoupling, independent of frequency, of the method for monitoring the electrode line from the HDT system. Because of the reciprocity of the electrode line, it is not possible, on the other hand, for signals fed in at the feed terminals in the push-pull mode to pass into the HDT system, as a result of which the method is independent of random switching states of the HDT system. In order to be able to feed a signal in push-pull mode into the electrode line, consisting of two lines, the short circuit for this mode must be rendered ineffective at the branch point. This could be accomplished, for example, by connecting a coil of large inductance into the electrode line in series in each case between the feed terminals and the branch point. Since currents of the order of magnitude of kA flow through the electrode line in monopolar operation, the two coils required for this would also need to be designed for these currents. - 7 - An advantageous refinement of the method envisages carrying out the feeding of the push-pull mode without such components as the coils mentioned. This is possible when the feed points are located at a predetermined spacing from the branch point, this spacing being dimensioned such that it corresponds approximately to a quarter of the line wavelength at the centre frequency of the generated unbalanced-to-earth pulse. At this frequency, the short circuit at the branch point is transformed into idling at the feed terminals and at adjacent frequencies this short circuit is transformed into a high-resistance reactance which is to be regarded as being connected in parallel at the feed terminals to the characteristic impedance of the line. A further advantage of this method consists in the fact that this monitoring method can be adapted automatically to the different operating conditions. This is achieved by virtue of the fact that an echo differential curve is generated as a function of a recorded actual echo curve and a stored, formed dynamic desired echo curve. By using a dynamic desired echo curve which varies in time, it is the case, for example, that seasonal influences on the electrode line are incorporated into the monitoring method, it thereby being possible for the case of a fault to be determined uniquely in each case. If a fault signal is generated, the monitoring method can be switched off. The generation of pulses is interrupted or switched off for this purpose. In an advantageous method, a predetermined static desired echo curve is enveloped by a tolerance band which is determined by limit curves running above and below this static desired echo curve. A formed dynamic desired echo curve is now checked with the aid of this static desired echo curve as to whether at least one amplitude of this dynamic desired echo curve is situated outside this tolerance band of the static desired echo curve. If this is the case at least once - 8 - inside a prescribed time interval, a fault signal is generated and the monitoring method is switched off. By using a predetermined static desired echo curve, it is possible to determine defects at the device for monitoring the electrode line which, if they occur in a creeping fashion, otherwise would fall under a temporally changing operating condition. In a further advantageous method, the dynamic desired echo curve is formed from a mean value of at least two temporally successive actual echo curves, that is to say, a mean value is always formed from a predetermined number of temporally successive actual echo curves and stored as a dynamic desired echo curve. As a result, with each new actual echo curve a new mean value is stored as a dynamic desired echo curve. This is performed, however, only when the evaluation of an echo differential curve has not generated a report signal - Further advantageous refinements of the method for monitoring an electrode line of a bipolar HDT system are specified in subclaims 5 to 10. By virtue of the fact that, in addition to a pulse-echo monitoring unit having a pulse generator and a receiving device, the device according to the invention for monitoring an electrode line of a bipolar HDT system has a feed device which is connected respectively on the output sxde to a feed terminal of the lines of the electrode line, a balanced-to-earth pulse is generated in push-pull mode from an unbalanced-to-earth pulse generated by the pulse generator. The pulse-echo monitoring unit is connected to "the inputs of the feed device. This feed device has on the input side a device for pulse conversion and on the output side two coupling capacitors which in each case connect the outputs of the device for pulse conversion to a feed terminal. Owing to the configuration of the feed device, on the one hand a balanced-to-earth pulse is generated in push-pull mode from an unbalanced-to-earth pulse of - 9 - the pulse generator, as a result of which the previously mentioned advantages of the push-pull mode by comparison with a common mode arise, and on the other hand disturbances coming from the HDT system are transmitted only very weakly damped to the receiving device. In an advantageous refinement of the feed circuit, there is provided as device for pulse conversion an isolating transformer with low-voltage and high-voltage windings, two coils and two arrestors, one coil and one arrestor in each case being connected electrically in parallel with a high-voltage winding. The tie point of the two high-voltage windings is connected to earth potential. The two coupling capacitors form with the two coils two high-pass filters which in each case are tuned to the centre frequency of the generated pulse. The arrestors protect the isolating transformer against overvoltages in the event of transient interference (lightening strike, switching impulse). In an advantageous device, the pulse generator has two voltage sources, two capacitors, two switches, two resistors and an operating device for the switches, each capacitor being connected in an electrically conducting fashion to a voltage source by means of a resistor. A tie point of these two capacitors and a tie point of the two voltage sources are connected in each case to earth potential. The capacitors can be connected to the output of the pulse generator by means of a switch in each case, the operating device being connected to a control output of the pulse generator. A narrow-band, square-wave pulse free from direct components and having a high spectral component at its centre frequency can be generated in a simple way with the aid of this pulse generator. It is also possible in principle to make use in the monitoring method of other pulse shapes which have the already mentioned spectral properties. For example, it is also possible to use a saw-tooth pulse running - 10 - symmetrically relative to the time axis. The generation of such a pulse is, however, more complicated. In a further advantageous embodiment, the receiving device has a device for real time recording of echo signals, an arithmetic unit, a main memory and input and output interfaces, the control input of this feed device being connected to a control input of the device for real time recording of echo signals. The arithmetic unit is connected to the main memory, the device for real time recording and to the interfaces. A signal input of the" device for real time recording is connected to the input of the receiving device, and a higher-order system controller being connected on the input and output sides to the output and input interfaces. Owing to the connection of the control output of the pulse generator to the control input of the device for real time recording, this device is triggered to emit pulses of the pulse generator. It is thereby possible to record the echo signals for a predetermined time, that is to say this part of the receiving device is operated online. The further processing of these recorded echo signals is performed offline, execution of this further processing being handled centrally in the arithmetic unit. In a further advantageous device, the feed terminals of the lines of the electrode line are arranged in each case at a spacing from a branch point of the electrode line, this spacing being, in particular, equal to a quarter of the wavelength of the free-space wavelength at the centre frequency of the pulse. Through selection of the spacing of these feed terminals from the branch point, there is no need for any sort of switching elements to be connected in series in the electrode line. For the centre frequency of the feed pulse, the short circuit at the branch point of the electrode line is transformed via the A/4-long line into idling at the feed point. Thus, for this frequency the A/4-long line together with the entire HDT system is not electrically available. In the case of this frequency, the pulse fed sees only the characteristic impedance of the two lines of the electrode line which lead to the earth electrode and to the branch point. At other frequencies, the short circuit at the branch point is transformed via the line, then no longer of length A/4, into a reactance which is to be regarded as connected in parallel, at the feed point to the characteristic impedance of the line. Owing to the use of conditions on the system side, and to the excitation of the push-pull mode, there is no need for any sort of additional switching measures to decouple the measuring arrangement from the station. It is thereby possible to dispense with the expensive dampers. Further advantageous refinements of the device for monitoring an electrode line of a bipolar HDT system are to be gathered from subclaims 13 to 19. For the purpose of a more detailed explanation . of the invention, reference is made to the accompanging which an exemplary embodiment of the device according to the invention is represented diagrammatically. Figure 1 shows a device according to the invention, having an electrode line of a bipolar HDT system, Figure 2 shows a block diagram of the pulse generator of the device according to Figure 1, Figure 3 shows a diagram of a generated pulse plotted against time t, Figure 4 shows a block diagram of the receiving device of the device according to Figure 1, Figure 5 shows a diagram of a recorded actual echo curve of a defective electrode line, plotted against time t, Figure 6 shows a diagram of an echo differential curve plotted against time t, in the case of a conductor-earth fault on the electrode line, and - 12 - Figure 7 shows a diagram of a static desired echo curve with associated tolerance band, plotted against time t. Figure 1 shows a device 2 according to the invention for monitoring an electrode line 4 of a bipolar high-voltage DC transmission system, of which for the sake of clarity only one converter station 6 is represented. In the case of a bipolar HDT system, which is also denoted as bipolar DC transmission, the two converter stations are interconnected by two DC lines 8 and 10, and each station has two converters 12 and 14 which are connected electrically in series by means of a connecting bus 16. The direct current is not led back through the earth in normal operation of this bipolar HDT system. For various reasons, inter alia in order to permit monopolar operation of the system in the case of a converter failure, the converter stations are also provided in bipolar DC transmissions with an earth connection which is connected to the station 6 on the connecting bus 16 by means of the electrode line 4. This electrode line 4 is insulated with respect to earth and normally comprises a line which is suspended on insulators. The electrode line 4 represented here is split up from the branch point 18, the so-called splitting point, into two lines 20 and 22 which are connected to earth potential at the end. These lines 20 and 22 of electrode line 4 can, if appropriate, be up to 100 km long. The second converter station, not represented here in more detail, of the bipolar HDT system is likewise equipped with an electrode line, that is to say the HDT system is of mirror symmetrical design. In fault-free operation - that is to say in balanced operation - virtually no current flows into this electrode line 4. Although the voltage between the electrode line and earth is normally small in relation to other voltages in the system, an earth fault, on the electrode line 4 poses the risk of injury to people or damage to otner system components. For this reason it - 13 - is necessary for earth faults, including high-resistance earth faults, to be capable of being discovered quickly and reliably. Moreover, it is important for the reliable operation of this bipolar HDT system also to detect the state of this electrode line 4. In the case of a line breakage and defective operation of the HDT system, unbalanced operation of this bipolar HDT system would then no longer be possible. The device 2 according to the invention is provided for the purpose of monitoring this electrode line 4 comprising two lines 20 and 22. This device 2 has a pulse-echo monitoring unit 24 and a feed device 26. This pulse-echo monitoring" unit 24 comprises a pulse generator 28 and a receiving device 30. The pulse generator 28, of which an advantageous embodiment is represented in more detail in Figure 2, is connected to input terminals of the feed device 26 by means of a coaxial cable 32. These terminals of the feed device 26 are, moreover, connected to input terminals of the receiving device 30. In order to ensure as little interference as possible in the signal, this coaxial cable 32 should be doubly shielded. Moreover, a control output of the pulse generator 2 8 is connected to a control input of the receiving device 30 by means of a control line 34. The feed device 2 6 is connected in each case on the output side to a feed terminal 36 and 38 of the lines 20 and 22 of the electrode line 4. These feed terminals 36 and 38 are arranged at a spacing from the branch point 18, this spacing a corresponding approximately to A/4, X representing the free-space wavelength at the centre frequency of a generated pulse u (t) of the pulse generator 2 8. Moreover, these feed terminals 36 and 38 are connected in each case to earth potential by means of an arrestor 40 and 42. These two arrestors 40 and 42 protect the feed device 26 on the high-voltage side against transient interference (lightening strike). - 14 - On the input side, the feed device 26 has a device 25 for pulse conversion, and on the output side two coupling capacitors 50 and 52. The device 25 for pulse conversion has an isolating transformer 44, two coils 46 and 48 and two arrestors 54 and 56. This isolating transformer 44 comprises two high-voltage windings 58 and 60 and a low-volt age winding 62. The tie point 64 of the two high-voltage windings 58 and 60 are connected to earth potential. The two terminals of the low-volt age winding 62 form the terminals of the feed device 26, to which the coaxial cable 32 is connected. The coil 46 or 48 is connected electrically in parallel with the high-volt age winding 58 or 60. Moreover, the arrestor 54 or .56 is connected electrically in parallel with the high-voltage winding 58 or 60 of the isolating transformer 44. The coupling capacitor 50 or 52 connects the feed point 36 or 38 to one terminal of the high-volt age winding 58 or 60. These two coupling capacitors 50 and 52 take over the coupling of the feed device 2 6 to the high-volt age potential of the electrode line 4. Consequently, these coupling capacitors 50 and 52 must be designed for the appropriate high-voltage level of electrode line 4. Together with the two coupling capacitors 50 and 52, the two coils 46 and 48 arranged in a balanced fashion relative to earth potential form a high-pass filter in each case. These high-pass filters block the low-frequency interference coming from the electrode line 4, that is to say characteristic current harmonics which are generated by the HDT system and also flow through the electrode line 4 during unbalanced operation of the HDT system. The arrestors 54 and 56 protect the high-voltage windings 58 and 60 of the isolating transformer 44 against overvoltages in the event of transient interference (lightening strike, switching impulse). These arrestors 54 and 56 are dimensioned for a much lower voltage than the arrestors 40 and 42. The isolating transformer 44 ensures matching of the impedance of the characteristic - 15 - impedance of the coaxial cable 32 to the characteristic impedance of the line 20 and 22 of the electrode line 4. Moreover, this isolating transformer 44 represents a balancing transformer which generates a balanced-to-earth pulse in push-pull mode from an unbalanced-to-earth pulse generated by the pulse generator. Figure 2 shows a block diagram of the pulse generator 28 of the pulse-echo monitoring unit 24 according to Figure 1. This pulse generator 28 has two voltage sources 66 and 68, two capacitors 70 and 72, two switches 7 4 and 76, two resistors 78 and 80 and an operating device 82 for the switches 7 4 and 76. The capacitor 70 or 72 is connected in an electrically conducting fashion to the voltage source 66 or 68 by means of the resistor 7 8 or 80. The tie point 84 of the two capacitors 70 and 72 is connected to the tie point 86 of the two voltage sources 66 and 68, which is also connected to earth potential. The charging current of the capacitors 70 and 72 is set with the aid of the resistors 78 and 80. These capacitors 70 and 72 can be connected in each case by means of the switches 74 and 76 to the output of the pulse generator 28, to which the coaxial cable 32 is connected. Electronic switches, for example transistors, are provided as the switches 74 and 76. The configuration of the operating device 82 depends on the selection of the electronic switches. Moreover, the operating frequency of the switches 74 and 76, and the charging cycles are mutually dependent. When the capacitors 70 and 72 are charged, at the instant t1 the switch 74 is firstly closed for a predetermined time interval t2-t1. After expiry of this time interval t2-ti, this switch 7 4 is opened and the switch 7 6 is simultaneously closed. After a further predetermined time interval t3-t2, this switch 76 is opened again. Thereafter, the two capacitors 70 and 74 are recharged by means of the voltage sources 66 and 68, in order to generate the next pulse u(t), as it is represented in Figure 3 in a diagram plotted against time t. When the first switch 7 4 is closed at the - 16 - instant t1, the operating device 80 sends a trigger signal ST, by means of the control line 34, to the control input of the receiving device 30, which is represented in more detail in Figure 4. The generated pulse u(t) in accordance with Figure 3 is symmetrical relative to the time axis t, that is to say it has no direct component. Moreover, this pulse u(t) has a marked spectral component at its centre frequency. The level of this centre frequency depends on whether, for example, data are transmitted on the electrode line 4, or whether this electrode line 4 is laid in the vicinity of power lines on which data transmission takes place additionally. Such data transmission generally takes place in a frequency band of, for example, 30 kHz to 500 kHz. When the pulse width of the generated pulse u(t) is selected to be appropriately narrow, its centre frequency is above 500 kHz. In the case of a pulse width corresponding to the reciprocal of the pulse duration t3-t1 of, for example 2 us, the centre frequency is 500 kHz. That is to say, the pulse width of the generated pulse u(t) should be smaller than 2 us. Since this pulse has only slight spectral components below its centre frequency, interference caused to data transmission devices is approximately zero. Other pulse shapes can also be used in principle. However, in the choice of other pulse shapes it should be ensured that as far as possible no direct component is present and that a marked spectral component is present in the case of a centre frequency. With these conditions, it is possible, in particular, to generate the pulse u(t) in accordance with Figure 3 with a high efficiency for a low outlay. The requirement that the pulse should as far as possible have no direct component is based on the fact that the isolating transformer 44 of the feed device 26 cannot transmit a direct component in the frequency spectrum of the pulse u(t). Figure 4 shows a block diagram of the receiving device 30 according to Figure 1. This receiving device 30 has a device 88 for real time recording of echo signals, an arithmetic unit 90, a main memory 92 and input and output interfaces 94 and 96. Moreover, this receiving device 30 also has a documentation memory 98 and a display screen 100. The input of the device 8 8 for real time recording of echo signals is connected to the input terminal of the receiving device 30, to which the coaxial cable 32 is connected. Moreover, a control terminal of this device 88 is connected to the control input of the receiving device 30, to which the control line 34 is connected. On the output side, this device 88 for real time recording is connected to the arithmetic unit 90, which is also connected to the main memory 92 and the documentation memory 98 in such a way that data can be exchanged. This arithmetic unit 90 is further connected on the input side to the input interface 94, in particular a binary input interface, and on the output side to the display screen 100 and the output interface 96, in particular a binary output interface. The receiving device 30 is connected by means of these two interfaces 94 and 96 to a higher-order system controller, which is not represented in more detail. The receiving device 30 receives operating and setting parameters from this system controller, which is a part of a control and protection system of the HDT system. A generated report signal or state signals pass to the control and protection system by means of the output interface 96. The device 88 for real time recording of echo signals comprises an analogue-to-digital converter and a memory, in particular a read/write memory which is connected downstream of the A/D converter. This A/D converter is started by means of the trigger pulse ST from the pulse generator 28, that is to say the A/D converter starts by digitizing the analogue input signals, that is to say the incoming echo signals. These digital samples are stored after the digitization. These two components of the device 88 operate online, that is to say the incoming echo - 18 - signals are processed in real time. The level of the sampling frequency of the A/D converter and the speed of the storage of the digital samples depends on how long the electrode line 4 is. This means that the length of the electrode line 4 determines the echo propagation time, and the period which is of interest for the evaluation is thereby fixed. Moreover, the level of the sampling frequency is also dependent on the memory capacity. These stored samples form, as time function, an actual echo curve EK in accordance with Figure 5. The further processing of this actual echo curve determined in real time now proceeds offline. For this purpose, these digitized samples are copied into the main memory 92. Moreover, these samples can be represented graphically on the display screen 100, that is to say the actual echo curve EK represented in Figure 5 appears on the display screen 100. This actual echo curve EK is compared by means of the arithmetic unit 90 with a dynamic desired echo curve stored in the main memory 92, that is to say an echo differential curve EDK is calculated, as is represented, for example, plotted against time t in a diagram in Figure 6. This echo differential curve EDK is provided with constant limit curves GKO and GKU running above and below this echo differential curve EDK. These two limit curves GKO and GKU thus form a tolerance band which is used to find points of faults. The actual echo curve EK in accordance with Figure 5 shows that in the case of no fault the pulse is reflected in a defined fashion at the end of the electrode line 4. The echo is cast back and is represented and evaluated at the receiver in the period. The result, for example, is this represented actual echo curve EK, which represents the pulse response of the entire system (coaxial cable 32, feed device 26 and defective electrode line 4). This actual echo curve EK represents, as it were, a fingerprint of the system without a fault. A typical actual echo curve - 19 - EK of an approximately 7.4 km long fault-free electrode line 4 is represented in Figure 5. The associated time axis t of this diagram is parameterized in kilometres of distance. It is possible to distinguish several regions in this actual echo curve EK. Those regions set forth below are enumerated: a) Feed pulse (1) b) Reflection at the isolating transformer 44 (2) c) Initial and transients of the feed device (3) d) Defined reflection from the end of the electrode line (4). If a fault now occurs on the electrode line 4 (phase-to-earth fault or line breakage), an additional echo of the point of the fault is produced. This leads to a change in the actual echo curve EK. In accordance with the representation of Figure 6, a phase-to-earth fault is approximately 4.5 km removed from the feed point 36, 38 and generates a distinct echo or an excursion LEF. At the same time, the defined actual echo curve EK from the end of the electrode line is also distorted, and this is illustrated in the case of the differential curve as a second excursion F2. The first excursion LEF, which is more closely situated in time, always originates from the point of a fault, and is to be used to determine the fault location if this is desired as a datum. In general, it is also possible to use the shape and/or the intensity of the echo reflected from the point of a fault to draw conclusions on the type of fault (phase-to-earth fault or line breakage). However, for permanently monitoring the electrode line 4 for faults, it suffices to monitor the echo differential curve EDK in a general way for excursions LEF which are situated outside the tolerance band. As already mentioned, a dynamic desired echo curve is used to determine the echo differential curve EDK. This desired echo curve is formed from at least two temporally successive recorded actual echo curves EK1 and EK2 by forming from these two curves EK1 and EK2 - 20 - a mean value echo curve which is then stored as desired echo curve. This calculation is continued dynamically, that is to say a new actual echo curve EK3 is used for calculating a new desired echo curve, a first actual echo curve EK1 no longer being used. Such a calculation can be carried out with the aid of a shift register, a new curve always being read in and the temporally oldest curve being read out. The mean average echo curve is calculated from the curves which are read into the shift register. Consequently, a new dynamic desired echo curve is calculated after each trigger pulse ST. This calculation is not begun until a comparison of a current actual echo curve with a current desired echo curve signals a state free from fault. It is established by means of setting parameters how many actual echo curves are to be used to calculate a dynamic desired echo curve. As an example, by means of this use of a dynamic desired echo curve which may vary temporally, seasonal influences on the electrode line 4 can be incorporated into the monitoring method, with the result that a case of a fault can be determined without ambiguity. In addition to the dynamic evaluation, a static evaluation also takes place. In this static evaluation, a fixed, temporally invariable desired echo curve EK*, which is assigned to a specific operating situation of the electrode line 4, is used. In accordance with the representation of Figure 7, this static desired echo curve EK* is enveloped by means of a tolerance band formed by means of limit curves GKOd and GKUd running above and below. In the case of the static evaluation, a formed dynamic desired echo curve is compared with the static desired echo curve EK* in such a way as to determine whether this dynamic desired echo curve still lies inside the formed tolerance band. If this does not apply once within a prescribed time interval, a fault signal is generated. As soon as a fault signal is generated, the device 2 for monitoring an electrode line 4 is switched off until the latter is reset again - 21 - manually. With the generation of a fault signal, the instantaneous dynamic desired echo curve and a number of preceding actual echo curves are buffered for documentation purposes in the documentation memory 98. Setting the tolerance band of the static desired echo curve EK* and setting the tolerance band of the echo differential curve are undertaken by means of setting parameters. A static desired echo curve EK* which belongs to a specific operating situation of the electrode line 4 is called by means of operating parameters. It is a precondition that a plurality of static desired echo curves EK* are stored in the main memory 92. -22- We Claim 1. Method for electrically monitoring an electrode line (4) of a bipolar high-voltage DC transmission system, this elcectrode line (4) being split up from a branch point (18) on into two lines (20, 22), having the following method steps: a) generating an electric pulse signal (u(t)) unbalanced with reference to earth potential, and emitting the generated pulse signal and a trigger signal (ST) when emitting the pulse signal, b) converting this emitted unbalanced pulse signal (u(t)) into a pulse signal balanced with reference to earth potential in push-pull mode, c) injecting the push-pull mode pulse signal into the two lines (20, 22) of the electrode line (4), d) forming a dynamic desired echo curve by recording an actual echo curve resulting from the injection of the push-pull mode pulse signal, e) forming an actual echo curve (EK) by real time recording of an echo signal resulting from the injection of a further push-pull mode pulse signal for a predetermined time, f) forming an echo difference curve (EDK) by forming the difference from the recorded actual echo curve (EK) and the dynamic desired echo curve, g) checking the echo difference curve (EDK) for amplitudes which project from a tolerance band formed by two predetermined constant limit curves (GKO, GKU) arranged symmetrically relative to the time axis, h) generating a fault signal as soon as at least one overshoot of the tolerance band is available, and i) switching off the pulse generation as soon as a fault signal is present. 2. Method according to Claim 1, the balanced-to-earth pulse in push-pull mode being fed into the two lines (20, 22) at a distance (a) of approximately 1/4 of a free-space wavelength (X) of the center frequency of the generated unbalanced-to-earth pulse (u(t)) from the branch point (18) of the electrode line (4). 3. Method according to Claim 1 or 2, a predetermined static desired echo curve (EK*) being enveloped by a tolerance band formed by a limit curve - 23 - (GKOd, GKUd) running above and below, a check being made, depending on this tolerance band of the static desired echo curve (EK*) , as to whether a formed dynamic desired echo curve is still arranged inside the tolerance band of this static desired echo curve (EK*) , a fault signal being generated as soon as an amplitude of the dynamic desired echo curve to be checked is outside this tolerance band at least once within a prescribed time interval, and the pulse generation being switched off as a consequence of the fault signal. 4. Method according to one of the previously mentioned Claims 1 to 3, it being the case that for the purpose of generating a dynamic desired echo curve a mean-value echo curve which is stored as a desired echo curve is formed continuously from at least two temporally successive actual echo curves (EK1, EK2, EK3) of an error-free operation of the electrode line (4). 5. Method according to one of Claims 1 to 4, the unbalanced-to-earth pulse (u (t) ) being generated periodically. 6. Method Recording to one of the previously mentioned Claims 1 to 5, a pulse with no direct components and of short pulse width is generated as the unbalanced-to-earth pulse (u(t)). 7. Method according to one of Claims 1 to 3, the instantaneous dynamic desired echo curve and the instantaneous recorded actual echo curve (EK) being stored for documentation purposes with the generation of a fault signal. 8. Method according to Claim 3, the tolerance band of the static desired echo curve (EK*) being generated as a function of predetermined operating parameters of the electrode line (4) . 9. Method according to one of the previously mentioned Claims 1 to 8, a predetermined number of recorded actual echo curves (EK1, EK2/ EK3) being stored continuously for documentation purposes. Method according to claim 1, 3 or 8, a static desired echo curve (EK*) with associated tolerance band in each being generated and stored for different operating conditions. Device (2) for monitoring an electrode line (4) of a bipolar high-voltage DC transmission system, this electrode line (4) being split up from a branch point (18) on into two lines (20, 22) this device (2) having a pulse-echo monitoring unit (24), which has a pulse generator (28) for generating an unbalanced to earth pulse (U (t)) and emitting a trigger signal (ST) during pulse emission and a receiving device (30), and a feed device (26), the pulse-echo monitoring unit (24) being connected on the output side to the input terminals of the feed device (26), the feed device (26) being connected respectively on the output side to a feed terminal (36, 38) of the lines (20, 22) of the electrode line (4), the feed device (25) for pulse conversion of the unbalanced to earth pulse (U(t)) into a balanced to earth pulse in push-pull mode and two coupling capacitors (50, 52) which connects the outputs of the device (25) for pulse conversion to the outputs of the feed device (26) for injecting the push-pull mode pulses into the two lines (20, 22). Device (2) according to claim 11, wherein said device (25) for pulse conversion is provided with an isolating transformer (44) which has one low-voltage and two high voltage windings (62, 58, 60) whose the point (64) is connected to earth potential. Device (92) according to claim 1, wherein said device (25) for pulse conversion is provided with an isolating transformer (44), two coils (46, 48) and two arrestors (54, 56) being provided as device (25) for pulse conversion, one coil (46, 48) and one arrestor (54, 56) in each case being connected electrically in parailei with a high voltage winding (56, 60). Device (2) according to one of claims 11 to 13, the pulse generator (28) having two voltage sources (66, 68), two capacitors (70, 72), two switches (74, 76), two resistors (78, 80) and an operating device (82) for the switches (74, 76), each capacitor (70, 72) being connected in an electrically conducting fashion to a voltage source (66, 68) by means of a resistor (78, 80), a tie point (84) of the two capacitors (70, 72), and a tie point (86) of the voltage sources - 25 - (66, 68) being connected in each case to earth potential, it being possible for the capacitors (70, 72} to be connected to the output of the pulse generator (28) by means of a switch (74, 76) in each case, and the operating device (82) being connected to a control output of the pulse generator (28). 15. Device (2) according to one of Claims 11 to 14, the receiving device (30) having a device (88) for real time recording of echo signals, an arithmetic unit (90), a main memory (92) and input and output interfaces (94, 96), the control input of the feed device (30) being connected to a control input of the device (88) for real time recording of echo signals, the arithmetic unit (90) being connected to the main memory (92) , the device (88) for real time recording and to the interfaces (94, 96), the signal input of the device (88) for real time recording being connected to the input of the receiving device (30), and a higher- order system controller being connected on the input and output sides to the output and input interfaces (96, 94) . 16. Device (2) according to one of the previously mentioned Claims 11 to 15, the feed terminals (36, 38) of the lines (20, 22) of the electrode line (4) being arranged in each case at a spacing (a) from the branch point (18) of the electrode line (4). 17. Device (2) according to Claim 16, the spacing (a) being equal to a quarter of the wavelength (X) of a free-space wavelength at the centre . frequency of the pulse {u (t) } . 18. Device (2) according to one of the previously mentioned Claims 11 to 17, the feed terminals (36, 38) being connected to earth potential by means of an arrestor (40, 42) in each case. 19. Device (2) according to one of Claims 11 to 18, the pulse generator (28) and the receiving device (30) forming a pulse-echo monitoring unit (24). - 26 - 20. Device (2) according to one of the previously mentioned Claims 11 to 19, a coaxial cable (32) being provided as connecting line between the pulse generator (28) and feed device (26) and between the feed device (26) and receiving device (30). 21. Device (2) according to Claim 20, the coaxial cable (32) being doubly shielded. 22. Device (2) according to Claim 15, the device (28) for real time recording having an analogue-to- digital converter with a downstream memory. 23. Device (2) according to Claim 15, the receiving device (30) having a documentation memory (98) which is connected to the arithmetic unit (90). 24. Device (2) according to Claim 15 or 23, the receiving device (3 0) having a display screen (100) which is connected on the input side to the arithmetic unit (90). Method and device for monitoring an electrode line of a bipolar high-voltage DC transmission system In order to monitor an electrode line (4), comprising two lines (20, 22), of a bipolar HDT system, a balanced-to-earth pulse is fed from an unbalanced-to-earth pulse (u(t)) in push-pull mode into the lines (20, 22), and an actual echo curve (EK) which is compared with a dynamic desired echo curve is recorded from the echo signals, a fault signal being generated upon overshooting of a tolerance band placed around the echo differential curve (EDK). The result is a monitoring method which is almost independent of the strongly fluctuating earth conductivity, which has a longer range in simultaneous conjunction with a lower dispersion of the echo, and which radiates less electromagnetic energy and requires no additional decoupling elements.

Documents:

00301-cal-1999-abstract.pdf

00301-cal-1999-claims.pdf

00301-cal-1999-correspondence.pdf

00301-cal-1999-description(complete).pdf

00301-cal-1999-drawings.pdf

00301-cal-1999-form-1.pdf

00301-cal-1999-form-2.pdf

00301-cal-1999-form-3.pdf

00301-cal-1999-form-5.pdf

00301-cal-1999-letters patent.pdf

00301-cal-1999-p.a.pdf

00301-cal-1999-priority document others.pdf

00301-cal-1999-priority document.pdf

301-CAL-1999-(26-10-2012)-FORM-27.pdf

301-CAL-1999-FORM-27.pdf