Title of Invention	INSTALLATION HAVING A DC VOLTAGE INTERMEDIATE CIRCUIT AND SELF-COMMUTATED CONVERTERS
Abstract	The invention relates to a method for regulating at least two inverters (1), which may be regulated either as rectifier or inverter, connected to each other by a DC link (4), for energy transmission and/or distribution, wherein a measured DC voltage (Udc_rl... Udc_rr; Udc_il... Udc_ii) and a measured DC current ( Idc_rl... Idc_rr; Idc_il... Idc_ii ) is measured at each inverter (1) and transmitted to a rectifier regulator (7_rr) for regulating the corresponding rectifier or to an inverter regulator (8_ii) for regulating the corresponding inverter, wherein each rectifier regulator (7_rr) and each inverter regulator (8_ii) give the difference between a given set DC voltage (Udco) and the relevant received measured DC voltage (Udc_rl... Udc_rr, Udc_il... Udc_ii) to give a differential DC voltage (du) and the difference between a set DC current (Idco_rl Idco_rr, Id-co_il Idco_n) and the corresponding received measured DC current (Idc_rl... Idc_rr, Idc_il... Idc_ii) to give a differential DC current (di), wherein the measured DC current, the measured DC voltage, the set DC current and the set DC voltage are normalised with which a regulation of inverters comprising switchable power semiconductors connected by a DC link can be carried out, wherein the proviso of set currents being zero is possible. According to the invention, each inverter is a self-commutated inverter with switchable power semiconductors and the rectifier regulation (7_rr) of the provided inverter (1) is regulated such that the sum of the product of the differential voltages (du) and the value of given set DC current ( \|Idco_r\|) at th corresponding rectifier and the differential current (di) is a minimum (du* \| Idco_r \| +di -- >Min) and the inverter regulation (8_ii) regulates the corresponding inverter (1) such that the sum between the differential voltage (du) and the differential current (di) is a minimum (du+di -->Min).

Title of Invention

INSTALLATION HAVING A DC VOLTAGE INTERMEDIATE CIRCUIT AND SELF-COMMUTATED CONVERTERS

Abstract

The invention relates to a method for regulating at least two inverters (1), which may be regulated either as rectifier or inverter, connected to each other by a DC link (4), for energy transmission and/or distribution, wherein a measured DC voltage (Udc_rl... Udc_rr; Udc_il... Udc_ii) and a measured DC current ( Idc_rl... Idc_rr; Idc_il... Idc_ii ) is measured at each inverter (1) and transmitted to a rectifier regulator (7_rr) for regulating the corresponding rectifier or to an inverter regulator (8_ii) for regulating the corresponding inverter, wherein each rectifier regulator (7_rr) and each inverter regulator (8_ii) give the difference between a given set DC voltage (Udco) and the relevant received measured DC voltage (Udc_rl... Udc_rr, Udc_il... Udc_ii) to give a differential DC voltage (du) and the difference between a set DC current (Idco_rl Idco_rr, Id-co_il Idco_n) and the corresponding received measured DC current (Idc_rl... Idc_rr, Idc_il... Idc_ii) to give a differential DC current (di), wherein the measured DC current, the measured DC voltage, the set DC current and the set DC voltage are normalised with which a regulation of inverters comprising switchable power semiconductors connected by a DC link can be carried out, wherein the proviso of set currents being zero is possible. According to the invention, each inverter is a self-commutated inverter with switchable power semiconductors and the rectifier regulation (7_rr) of the provided inverter (1) is regulated such that the sum of the product of the differential voltages (du) and the value of given set DC current ( |Idco_r|) at th corresponding rectifier and the differential current (di) is a minimum (du* | Idco_r | +di -- >Min) and the inverter regulation (8_ii) regulates the corresponding inverter (1) such that the sum between the differential voltage (du) and the differential current (di) is a minimum (du+di -->Min).

Full Text	Description Installation having a DC voltage intermediate circuit and self- commutated converters The invention relates to a method for closed-loop control of at least two converters, which can be controlled as rectifiers or inverters and are connected to one another via a DC voltage link, in the field of power transmission and/or distribution, in which a measurement DC voltage (Udc_r1...Udc_rr; Udc_i1...Udc_ii) and a measurement direct current (Idc_r1...Idc_rr; Idc_i1...Idc_ii) are in each case measured at each converter and are transmitted to a rectifier regulator for closed-loop control of the respective rectifier, or are transmitted to an inverter regulator for closed-loop control of the respectively associated inverter, wherein each rectifier regulator and each inverter regulator in each case forms the difference between a predetermined nominal DC voltage (Udco) and the respectively received measurement DC voltage (Udc_r1...Udc_rr; Udc_i1...Udc_ii), producing a difference DC voltage (du) and, furthermore, the difference between a nominal direct current (Idco_r1...Idco_rr, Idco_i1...Idco_ii) and the respectively received measurement direct current (Idc_r1...Idc_rr; Idc_i1...Idc_ii) , producing a difference direct current (di), where the measurement direct current, the measurement DC voltage, the nominal direct current and the nominal DC voltage are in a normalized form. A method such as this is already known from WO 2007/033620 Al, which describes a closed-loop control method for high-voltage direct-current transmission in which electrical power is transmitted between AC voltage systems via a direct-current circuit. The high-voltage direct-current transmission (HVDCT) installation which is used for power transmission consists of a rectifier and an inverter, which are connected to one another via a direct current link. Transformers are provided in order to couple the converters to the respectively associated AC voltage system. For closed-loop control of the rectifier or inverter, a measurement DC voltage Udc_r or Udc_i, and a respective measurement direct current, Idc_r or Idc_i respectively, are recorded and are transmitted to the respective closed-loop control system both at the rectifier and at the inverter. A nominal DC voltage Udco and a nominal direct current Idco are determined from the predetermined power which is intended to be transmitted, with the aid of a function transmitter. The difference DC voltage du is then calculated from the difference between the nominal DC voltage Udco and the measurement DC voltage Udc_r, Udc_i. The difference direct current di is correspondingly formed from the difference between the nominal direct current Idco and the respectively determined measurement direct current Idc_r and Idc_i. All the values are in this case in normalized form, with the normalization process being carried out, for example, with respect to a rated DC voltage and a rated direct current, or else with respect to a nominal direct current and the nominal DC voltage. The rectifier regulator now controls the rectifier in such a way that the sum of the difference DC voltage du and of the difference direct current di is a minimum. The inverter regulator in contrast controls the inverter such that the difference between the difference direct current di and the difference DC voltage du is a minimum. The already known methods are, however, suitable only for so-called externally commutated converters, in which, for example, thyristors are used which cannot be turned off by means of trigger signals. Converters such as these allow current to flow in only one direction in the direct current circuit. The power flow can be reversed only by reversing the polarity of the voltage across the respective converter. Furthermore, in this installation, which is referred to in the following text as a traditional HVDCT installation, the operating current must always be greater than 0.05 p.u., because of the need to avoid intermittent currents in the current nominal value. This allows the abovementioned normalization with respect to the nominal values. As a result of the progressive improvement in the field of power electronics, it has become possible to also use power semiconductors which can be turned off, for example IGBTs or GTOs, for converters in the field of power transmission, and in particular for HVDCT. A converter such as this, which is also called a voltage source converter (VSC), is connected to a further VSC via a DC voltage intermediate circuit. Each power semiconductor which can be turned off in the converter has a freewheeling diode connected in parallel with it. The power flow is no longer reversed by reversing the polarity of the voltage on the respective VSC but by reversing the current flowing via the respective VSC. Furthermore, direct current nominal values equal to zero can be used for the respective VSC. It is therefore impossible to use the method of this generic type for controlling a VSC in an HVDCT installation. WO 2007/033619 A1 discloses a closed-loop control method for direct current transmission having a plurality of converters, in which the method mentioned above is likewise used. In this case as well, nominal currents of more than 0.05 p.u. are required, in particular at the rectifier, in order to avoid intermittent currents. The object of the invention is to provide a method of the type mentioned initially, which allows closed-loop control to be carried out for converters which have power semiconductors which can be turned off and are connected to one another via a DC voltage intermediate circuit. while at the same time making it possible to preset nominal currents equal to zero. The invention achieves this object in that each converter is a self-commutated converter (1) having power semiconductors which can be turned off, and the rectifier regulator controls the respectively associated converter such that the sum of the product of the difference DC voltage (du) and the magnitude of the nominal direct current (\|Idco_r\|), which is provided at the respectively associated rectifier, and the difference direct current (di) is a minimum (du\|Idco_r\|+di—>Min) and the inverter regulator controls the respectively associated converter such that the sum between the difference DC voltage (du) and the difference direct current (di) is a minimum (du+di—>Min). The invention transfers the closed-loop control method known from traditional HVDCT to HVDCT installations in which VSCs and a DC voltage intermediate circuit are used. In the case of these so-called voltage source converters (VSC), the voltage polarity can no longer be reversed because of the DC voltage intermediate circuit, by which means the power flow is reversed in a conventional HVDCT. In converters (VSCs) such as these, the power flow is reversed by reversing the current flow. These physical differences are taken into account in the closed-loop control method according to the invention. Furthermore, the method according to the invention also makes it possible to preset nominal currents with a value of zero. This was impossible in the case of the known closed-loop control methods for conventional HVDCT. The method according to the invention therefore provides simple and flexible closed-loop control for two or more VSCs which are connected to one another via a DC voltage link. The DC voltage link has a polarity which does not change during operation of the HVDCT installation, within the scope of the invention, even when the power flow is reversed. In this case, for example, the DC voltage link is a bipolar DC voltage link which extends between two converters. According to this variant, the HVDCT installation to be controlled consists of a rectifier, which is connected via an inductance, for example a transformer, to an AC voltage system which provides power. The inverter, which is connected to said rectifier via the bipolar DC voltage link, is likewise connected via an inductance, for example a transformer, to a second AC voltage system, which, for example, has the loads to be supplied. However, in contrast to this, a plurality of converters can also be controlled using the method according to the invention, wherein the converters, or in this case the VSCs, are connected to one another via a DC voltage link with any desired topology. The method according to the invention is completely independent of the topology of the DC voltage link. Expediently, the nominal DC voltage Udco of each rectifier regulator and the nominal DC voltage Udco of each inverter regulator are identical. Complex definition of the nominal DC voltage at each converter station, with rapid and reliable telecommunication for signaling the respective nominal DC voltage to other stations, has therefore become superfluous within the scope of the invention. Expediently, the converters are at a distance of at least 1 kilometer from one another. All HVDCT installations which use the so-called back-to-back configuration, in which inverters and converters are installed physically alongside one another and are used only for coupling different AC voltage systems in a manner which can be subjected to closed-loop control, are excluded from this expedient further development. The apparatus for which the method according to this advantageous further development is designed is therefore used to transmit electrical power over a relatively long distance. Advantageously, each measurement DC voltage Udc_rr or Udc_ii and the nominal DC voltage Udco are normalized with respect to the nominal DC voltage Udco. This allows stable closed-loop control, in particular when the power levels to be transmitted are low. This normalization, which is already known from conventional HVDCT, is, however, carried out for the purposes of the exemplary embodiment according to the invention only in conjunction with the respective measurement and nominal DC voltage. In contrast, the nominal direct current is not normalized, in order to allow the use of nominal direct currents equal to zero in this expedient further development of the invention as well. Advantageously, two converters are provided, one of which is operated as a rectifier and the other converter is operated as an inverter, wherein a DC voltage intermediate circuit extends between the rectifier and the inverter. According to this refinement, the method is used for an HVDCT installation with VSCs and a DC voltage intermediate circuit without any branches. For example, the DC voltage intermediate circuit is designed with one pole, using the ground as the return conductor. In contrast to this, the DC voltage link has two poles, with the positive pole and negative pole of the DC voltage link being in the form of cable conductors. According to one refinement, which differs from this, the method according to the invention is provided for at least three converters, between which a DC voltage system extends. The DC voltage system may have any desired topology. According to one expedient further development in this context, a respectively associated rectifier nominal DC power (Pdco_r1...Pdco_rr) or inverter nominal DC power (Pdco_i1...Pdco_ii) is in each case defined for each rectifier regulator and for each inverter regulator, respectively, wherein the sum of all the rectifier nominal DC powers and all the inverter nominal DC powers is equal to zero. By way of example, the respective nominal DC power levels are defined by a central control center and are transmitted from there to the respective converter regulators. There is no need whatsoever for a powerful communication link for transmission, for example as is the case with a conventional HVDCT. For the purposes of the invention, it is sufficient to transmit the respective nominal DC power levels via the Internet, or any other simple, and therefore cost-effective, form of telecommunication to the respective regulator. Expediently, the nominal direct current respectively associated with the converter is determined at each converter from the nominal DC power associated with it and from the nominal DC voltage, which is common to all of them. According to this advantageous further development, the method according to the invention is simple and clear. Expediently, the closed-loop control of each rectifier and each inverter is carried out over the entire operating range of the rectifier or of the inverter both on the basis of the respectively associated rectifier difference direct current (di_r1...di_rr) and on the basis of the rectifier difference DC voltage (du_r1...du_rr) and, respectively, on the basis of the respectively associated inverter difference direct current (di_r1...di_rr) and on the basis of the associated inverter difference DC voltage (du_i1...du_ii). Advantageously, a respective measurement DC voltage (Udc_r1...Udc_rr; Udc_i1...Udc_ii) and a respective measurement direct current (Idc_r1...Idc_rr; Idc_i1...Idc_ii) are measured at each converter and are transmitted to a rectifier regulator or to an inverter regulator. According to one expedient further development in this context, a number of converters are controlled such that a direct current which flows through an isolating switch which is arranged in the DC voltage link is equal to zero. According to this advantageous further development, simple isolating switches can be used in the DC voltage link, since switching can always take place when no current is flowing. Complex resonant circuits which apply a zero crossing to the current by means of the switch, thus extinguishing any arc which is struck on switching, have likewise been rendered superfluous by this advantageous refinement of the invention. According to one modification in this context, at least one converter controls the respective measurement direct current recorded at it at zero, and at least one isolating switch, which is arranged in the DC voltage link, is then opened. Zero current regulation is carried out by using a nominal direct current equal to zero. According to one preferred refinement of this further development, all the converters which are connected to one another by a DC voltage link are controlled to zero. In other words, a nominal direct current zero is used for each regulator. According to one expedient further development in this context, a switching-off time is set and is transmitted to the respectively affected regulators, wherein said regulators control the measurement direct current associated with them to be zero when the switching-off time is reached. According to this expedient further development, all the regulator units have a time signal transmitter, for example an accurate clock, which provides a common, or substantially common, time signal for all the regulators. This time signal is compared with the set and transmitted switching-off time. If the difference between the switching-off time and the measured time is less than a threshold value set in advance, that is to say when the switching-off time is reached, the respective regulating unit presets a nominal direct current of zero, such that the measurement direct current is controlled at zero. This refinement ensures that all the affected regulators which, for example, are selected as such by a central control station, or all the regulators, are transferred to zero-current control at the same time. According to one expedient further development in this context, the measurement direct current is controlled to be zero by each affected converter over a zero-current time period, and the normal closed-loop control method is then restarted, that is to say once the zero-current time period has elapsed. Within the zero-current time period which, for example, is in the region of a few milliseconds, the desired isolating switch or switches in the DC voltage link is or are now opened. This makes it possible, in a simple manner, to deliberately disconnect an area or a converter from the DC voltage system with any desired topology, without likewise having to interrupt the power transmission via those links or converters which have not been turned off. In fact, within the scope of the invention, it is possible to disconnect a section of the DC voltage link or a converter specifically from the DC voltage system by rapidly reducing the direct current. The normal closed-loop control of the HVDCT installation is then started again. The closed-loop control method according to the invention then automatically approaches the regulator operating points required for this purpose, without any knowledge of the new topology of the DC voltage link. Complex data transmissions, calculations or the like have become superfluous as a result of the invention. Further expedient refinements and advantages of the invention are the subject matter of the following description of exemplary embodiments of the invention, with reference to the figures of the drawing, in which the same reference symbols refer to components having the same effect, and in which: Figure 1 shows an HVDCT installation having a plurality of voltage source converters, which are connected to one another via a DC voltage system, Figure 2 shows parameters and controlled variables used by one exemplary embodiment of the closed-loop control method according to the invention, Figure 3 shows one exemplary embodiment of the method according to the invention for a plurality of VSCs, which are linked to one another via a DC voltage system, and Figure 4 shows one exemplary embodiment of the method according to the invention, in which one specific section of the DC voltage link is disconnected from the DC voltage system by means of a single isolating switch. Figure 1 schematically illustrates an HVDCT installation having a multiplicity of converters 1, which are each connected via a transformer 2 to an AC voltage system 3. Each converter 1 is a so-called voltage source converter (VSC), having power semiconductors which can be turned off, such as IGBTs or GTOs, each of which has a freewheeling diode connected back-to-back in parallel with it. In this case, each converter 1 is connected to a DC voltage system 4 as a DC voltage link, which DC voltage system 4 may have any desired topology. Furthermore, each VSC 1 has an associated smoothing inductor 5. A VSC 1, a transformer 2, and a smoothing inductor 5 are part of a rectifier station or inverter station, depending on how the respective VSC is controlled by the regulating unit or regulator, which is not illustrated in the figure. In Figure 1, those rectifier stations which have VSCs operated at rectifiers are annotated r1, r2, r3 ... rr, whereas those inverter stations with VSCs which are operated as inverters are annotated i1, i2, i3 ... ii. Figure 1 also shows that each converter station rr and each inverter station ii has sensors for recording the measurement DC voltage Udc_rr at this station as well as the direct current Idc_rr or Idc_ii flowing there at the VSC 1. As can likewise been seen from Figure 1, the respective AC voltage systems 3 are also connected to one another via AC voltage links 6. The alternating current link 6 does not adversely affect the method according to the invention for closed-loop control of the VSC. Figure 2 is a graph, in which the normalized direct current is shown on the abscissa and the normalized DC voltage is shown on the ordinate. This is based on the assumption that, in contrast to the illustration in Figure 1, only one rectifier is connected via a DC voltage intermediate circuit to a VSC operating as an inverter. In the graph, the parameters and controlled variables of the rectifier are shown in the left-hand quadrant of the graph, and corresponding variables for the inverter are shown in the right-hand quadrant. The measurement direct current and the measurement DC voltage at the measurement point X_r are plotted on the rectifier side, whose measured and nominal values are annotated _r. The measurement DC voltage Udc_r is therefore greater than the nominal DC voltage Udco or Udco_r. If the difference DC voltage du_r is formed and normalized with respect to the nominal DC voltage Udco, this results in . This value must be less than zero for power to be transmitted to the inverter. The nominal direct current Idco_r of the rectifier is, by- definition, negative, as a result of which: Idco_r=-Idco. The difference direct current di, which is formed from the difference between the nominal and the measurement direct current, is then given by: di_r=-Idco_r+Idc_r. This is not normalized with respect to the nominal direct current Idco_r, in order to allow a nominal direct current of zero for closed- loop control. The closed-loop control is now carried out by minimizing the rectifier control error de_r=du_r\|Idco_r\|+di_r being, that is to say becoming zero. The straight line shown on the left is therefore a tangent to the hyperbola which is shown by a dashed line and represents the invariant nominal DC power Pdco_r=-Pdco. Since the difference DC voltage du_r is multiplied by the magnitude of the nominal direct current \|Idco_r\| in order to calculate the control error de_r, this results in a stable, rapidly active control response of the rectifier, because the closed-loop voltage control is suppressed when the nominal direct currents are small. Closed- loop voltage and current control are carried out, simultaneously and in a substantially equivalent form to one another, only for higher nominal direct currents Idco_r. The measurement direct current Idc_i and the measurement DC voltage Udc_i at the operating point X_i are plotted on the inverter side. As on the rectifier side, the difference DC voltage du_i and the difference direct current di_i are calculated, with the difference voltage du_i once again being normalized with respect to the nominal DC voltage Udco. du_i must be greater than zero for the desired power transmission. Normalization with respect to the nominal direct current in order to form the difference direct current di_i is also not carried out in this case. It is therefore also possible to preset nominal DC currents Idco_i equal to 0. For the desired power transmission, the difference direct current di_i is less than zero. The control error de_i is formed from the sum of the difference DC voltage du_i and the difference direct current di_i. The closed-loop control now attempts to minimize the control error de_i to be zero. On the graph shown in Figure 2, this results in a straight line which intersects the hyperbola of the invariant DC power Pdco_i at the point W_i. The straight line de_i is reminiscent of the response of a pure resistance, as a result of which the inverter regulator can also be referred to as a resistance regulator. Figure 3 schematically illustrates one exemplary embodiment of the method according to the invention for an HVDCT installation as shown in Figure 1, in which the parameters and variables shown in Figure 2 are used for a rectifier regulator 7_r1 for the rectifier station r_1, and for an inverter regulator 8_i1 for the inverter station i1. The rectifier regulators 7_rr, which are not shown in the figure, and the inverter regulator 8_ii, which is not illustrated, are physically identical. As can be seen from Figure 3, the rectifier regulator 7_r1, and therefore each rectifier regulator, receives a nominal DC power Pdco_r1 assigned to it. A corresponding situation applies to the inverter regulator 8_i1, in which case the respectively associated nominal DC power is transmitted from a central control station to the respective stations via communication links which are not shown, for example a simple radio link. In this case, the sum of all the nominal DC powers is equal to zero: ?Pdco_rr+?Pdco_ii=0. The invention makes complex, rapid and reliable transmission, as is the case for conventional HVDCT, superfluous. The received nominal DC power Pdco_r1 and Pdco_i1 is in each case supplied to a divisor 9, to whose second input the nominal DC voltage Udco, which is the same for all the stations, is applied. By way of example, the nominal DC power is likewise transmitted from a central station. The divider 9 forms the quotient of the respective nominal DC power Pdco and the nominal DC voltage Udco, producing the nominal direct current Idco_r1 or Idco_i1, with the respective nominal direct current being supplied to a limiter 10, which limits the nominal direct current Idco_r1 or Idco_i1 to a respective minimum nominal direct current Imin_r1,Imin_i1 and a respective maximum nominal direct current Imax_r1,Imax_i1. The nominal direct current Idco_r1 or Idco_i1 is then supplied to an adder 11, which in each case forms the difference from the nominal direct current Idco_r1 or Idco_i1 and the measurement direct current Idco_r1 or Idco_i1. The difference direct current di_r1 obtained in this way is then supplied to a further adder 11. The value at the second input of said adder 11 is derived from the measured voltage and the nominal voltage. For this purpose, the divisor 12 in each case determines the quotient of the measurement DC voltage Udc_r1 or Udc_i1 and the nominal DC voltage Udco. The measurement DC voltage Udc_r1 or Udc_i1 which have been normalized in this way is then subtracted from 1, producing the respective difference DC voltage du_r1 or du_i1. As stated above, for the inverter regulator 8_il, the difference DC voltage du_i1 obtained in this way is supplied to the second input of the adder 11 which, by addition of its inputs, calculates the inverter control error de_i, which is then supplied to a regulator 13. In contrast to the inverter regulator 8_i1, in the case of the rectifier regulator 7_r1,the difference DC voltage du_r1 is multiplied by the magnitude of the nominal direct current \|Idco_r1\|. An absolute-value generator 23 is provided in order to form the magnitude \|Idco_r1\| from Idco_r1,with the product \|Idco_r1\| * du_r1 being formed by means of a multiplier 24. In the case of the rectifier regulator 7_r1,the difference direct current di_r1 is added to said product of du_r1 and \|Idco_r1\| by means of the adder 11, in order to produce the control error de_r of the rectifier. The control error of the rectifier de_r1 is then supplied to the regulator 13 and, finally, to a module management system 14, which in this case provides open-loop control for the power semiconductors in the respective VSCs. The VSCs illustrated in Figure 3 are so-called multilevel VSCs which, like all VSCs, consist of a bridge circuit of electrical valves. However, in the case of multilevel VSCs, each electrical valve is formed from a series circuit of bipolar submodules which each have an energy store and, in parallel with the energy store, a circuit composed of power semiconductors, such that the voltage across the energy store, or else a zero voltage, is dropped across the respective submodule depending on the operation of the power semiconductors. The voltage which is dropped in total across the electrical valves is additively composed of the output voltages of the submodules, and can therefore be changed in steps. In order to operate the power semiconductors in the submodules, the output of the regulator 13 is connected to the input of the so-called module management system, whose detailed configuration will not be described for the purposes of the invention, since it is not essential to the invention. The module management system operates the power semiconductors in the submodules corresponding to the output of the regulator 13. However, within the scope of the invention, it is also possible for the regulator 13 to be followed by a pulse width modulator rather than by a module management system, which pulse width modulator is designed for open-loop control of two-stage or three-stage voltage source converters. Figure 4 likewise shows an HVDCT installation, in order to illustrate one exemplary embodiment of the method according to the invention. In this case as well, a plurality of rectifier stations r1, r2,...rr and a plurality of inverter stations i1, i2...ii are once again connected to one another via a DC voltage system 4 of any desired topology. In addition to the illustration shown in Figures 3 and 1, each rectifier station rr and each inverter station ii has a protection unit 15 in addition to a rectifier regulator 7_rr or inverter regulator 8_ii, respectively, which protection unit 15 is connected to the respective rectifier regulator 7_rr or inverter regulator 8_ii. The respective rectifier regulator 7_rr or inverter regulator 8_ii is also in each case connected to a protection device 16 of a circuit breaker 17, with the switch 17 being arranged between the AC voltage system 3 and the transformer 2 . In the event of a fault, for example high short-circuit currents, it is possible to disconnect each VSC from the respective AC system 3 by means of the multi-pole switch 17 on the AC voltage side which, for example, is a circuit breaker for switching high short-circuit currents. In this case, protection tripping takes place via the protection unit 15, which causes the respective regulating unit 7 to emit a tripping command to the protection device 16. The protection device 16 opens the circuit breaker 17 when a tripping command occurs. In order to allow DC voltage sections 18 to also be turned off or disconnected within the DC voltage link 4, DC voltages switches 19 are arranged in the DC voltage links. A DC voltage protection unit 2 0 is used to trip the DC voltage switch or switches 19 and is connected to the output of a DC voltage current sensor 21, which produces direct current values Idc_bl or Idc_b2, corresponding to the direct current flow via the respective switch 19. The DC voltage protection unit 2 0 is in turn connected to the protection device 16 for the DC voltage switch 19. The arrow 22 shown in Figure 4 indicates a ground fault, as a result of which high short-circuit currents flow in the DC voltage section 18. If the direct current values Idc_bl and Idc_b2 which are supplied to the respective DC voltage protection unit 2 0 exceed a previously defined threshold value, or some other criterion, the respective DC voltage protection unit 20 defines a switching-off time toff in the near future, and transmits the switching-of f time toff to the respective rectifier regulator 7_r1,7_r2, 7_rr or inverter regulator 8_i1, 8_i2, 8_ii. These regulators are connected to a timer, such that all the regulators can determine that the switching-off time has been reached, approximately at the same time. The protection unit 15 now operates the respective regulating unit such that this in each case sets the respective nominal direct current Idco_r1,Idco_r2, Idco_rr or Idco_i1, Idco_i2 and Idco_ii to zero throughout a zero-current time period, such that the respective measurement direct current is controlled at zero. Therefore, the respective direct current Idc_b1 or Idc_b2 flowing through the direct current switches 19 is also equal to zero. The switch 19 can now be opened with no current flowing. The DC voltage protection unit 20 likewise identifies that the switching-off time has been reached, by time comparison. After a safety time interval has passed, which is shorter than a zero-current time period, the DC voltage switch 19 is opened with no current flowing, thus disconnecting the DC voltage section 18. The normal closed-loop control method is started again once the zero-current time period has elapsed. For the purposes of the invention, the respective regulators need not be informed of the change in the topology of the DC voltage link 4. The closed-loop control system automatically changes to stable closed-loop control points without any further additional effects. This allows the DC voltage system 4 to be switched with little complexity. The invention renders parallel resonant circuits, as used for DC voltage switches according to the prior art, superfluous. WE CLAIM 1. A method for closed-loop control of at least two converters (1), which can be controlled as rectifiers or inverters and are connected to one another via a DC voltage link (4), in the field of power transmission and/or distribution, in which a measurement DC voltage (Udc_r1...Udc_rr; Udc_i1...Udc_ii) and a measurement direct current (Idc_r1...Idc_rr; Idc_i1...Idc_ii) are in each case measured at each converter (1) and are transmitted to a rectifier regulator (7_rr) for closed-loop control of the respective rectifier, or are transmitted to an inverter regulator (8_ii) for closed-loop control of the respectively associated inverter, wherein each rectifier regulator (7_rr) and each inverter regulator (8_ii) in each case forms the difference between a predetermined nominal DC voltage (Udco) and the respectively received measurement DC voltage (Udc_r1...Udc_rr; Udc_.i1...Udc_ii), producing a difference DC voltage (du) and, furthermore, the difference between a nominal direct current (Idco_r1...Idco_rr, Idco_i1...Idco_ii) and the respectively received measurement direct current (Idc_r1...Idc_rr; Idc_i1... Idc_ii), producing a difference direct current (di), where the measurement direct current, the measurement DC voltage, the nominal direct current and the nominal DC voltage are in a normalized form, characterized in that - each converter is a self-commutated converter (1) having power semiconductors which can be turned off, and the rectifier regulator (7_rr) controls the respectively associated converter (1) such that the sum of the product of the difference DC voltage (du) and the magnitude of the nominal direct current (\|Idco_r\|) , which is provided at the respectively- associated rectifier, and the difference direct current (di) is a minimum (du\|Idco_r\|+di?Min) and the inverter regulator (B_ii) controls the respectively- associated converter (1) such that the sum between the difference DC voltage (du) and the difference direct current (di) is a minimum (du+di?Min). 2. The method as claimed in claim 1, characterized in that the nominal DC voltage (Udco) of each rectifier regulator and the nominal DC voltage (Udco) of each inverter regulator are identical. 3. The method as claimed in claim 1 or 2, characterized in that the converters (1) are installed at a distance of at least 1 km from one another. 4. The method as claimed in one of the preceding claims, characterized in that each measurement DC voltage (Udc_rr) and the nominal DC voltage (Udco) are normalized with respect to the nominal DC voltage (Udco). 5. The method as claimed in one of the preceding claims, characterized in that two converters (1) are provided, one of which is operated as a rectifier and the other converter (1) is operated as an inverter, wherein a DC voltage intermediate circuit (4) extends between the rectifier and the inverter. 6. The method as claimed in one of claims 1 to 4, characterized in that at least three converters (1) are provided, between which a DC voltage system (4) extends. 7. The method as claimed in claim 6, characterized in that a respectively associated rectifier nominal DC power (Pdco_r1...Pdco_rr) or inverter nominal DC power (Pdco_i1...Pdco_ii) is defined for each rectifier regulator (7_rr) and for each inverter regulator (8_ii), respectively, wherein the sum of all the rectifier nominal DC powers and all the inverter nominal DC powers (?Pco_rr+?Pdco_ii) is equal to zero. 8. The method as claimed in claim 7, characterized in that the nominal direct current (Idco_rr; Idco_ii) associated with the respective converter (1) is determined at each converter (1) from the nominal DC power (Pdco_rr; Pdco_ii) associated with it and from the nominal DC voltage (Udco), which is common to all of them. 9. The method as claimed in one of the preceding claims, characterized in that the closed-loop control of each rectifier and each inverter is carried out over the entire operating range of the rectifier or of the inverter both on the basis of the respectively associated rectifier difference direct current (di_r1...di_rr) and on the basis of the rectifier difference DC voltage (du_r1...du_rr) and, respectively, on the basis of the respectively associated inverter difference direct current (di_rl...di_rr) and on the basis of the associated inverter difference DC voltage (du_i1...du_ii). 10. The method as claimed in one of the preceding claims, characterized in that a respective measurement DC voltage (Udc_r1...Udc_rr; Udc_i1...Udc_ii) and a respective measurement direct current (Idc_rl. . . Idc_rr; Idc_.i1...Idc_ii) are measured at each converter (1) and are transmitted to a rectifier regulator (7_rr) or to an inverter regulator (8_ii). 11. The method as claimed in claim 10, characterized in that a number of converters (1) are controlled such that a direct current (Idc_b1;Idc_b2) which flows through an isolating switch which is arranged in the DC voltage link (4) is equal to zero. 12. The method as claimed in claim 10 or 11, characterized in that at least one converter (1) controls the respective measurement direct current (Idc_rr;Idc_ii) recorded at it at zero, and at least one isolating switch (19), which is arranged in the DC voltage link (4), is then opened. 13. The method as claimed in claim 12, characterized in that a switching-off time (toff) is set and is transmitted to the respectively affected regulators, wherein said regulators control the measurement direct current (Idc_rr; Idc_ii) associated with them to be zero when the switching-off time (toff) is reached. 14. The method as claimed in one of claims 11 to 13, characterized in that the measurement direct current (Idc_rr;Idc_ii) is controlled to be zero by each affected converter (1) over a zero-current time period, and the normal closed-loop control method is then started. 15. The method as claimed in one of claims 11 to 14, characterized in that one of the converters (1) is disconnected from the DC voltage link (4) by opening the isolating switches (19). 16. The method as claimed in one of claims 11 to 15, characterized in that a section (18) of the DC voltage link (4) is selectively- disconnected from the DC voltage link (4) by switching the isolating switches (19). The invention relates to a method for regulating at least two inverters (1), which may be regulated either as rectifier or inverter, connected to each other by a DC link (4), for energy transmission and/or distribution, wherein a measured DC voltage (Udc_rl... Udc_rr; Udc_il... Udc_ii) and a measured DC current ( Idc_rl... Idc_rr; Idc_il... Idc_ii ) is measured at each inverter (1) and transmitted to a rectifier regulator (7_rr) for regulating the corresponding rectifier or to an inverter regulator (8_ii) for regulating the corresponding inverter, wherein each rectifier regulator (7_rr) and each inverter regulator (8_ii) give the difference between a given set DC voltage (Udco) and the relevant received measured DC voltage (Udc_rl... Udc_rr, Udc_il... Udc_ii) to give a differential DC voltage (du) and the difference between a set DC current (Idco_rl Idco_rr, Id-co_il Idco_n) and the corresponding received measured DC current (Idc_rl... Idc_rr, Idc_il... Idc_ii) to give a differential DC current (di), wherein the measured DC current, the measured DC voltage, the set DC current and the set DC voltage are normalised with which a regulation of inverters comprising switchable power semiconductors connected by a DC link can be carried out, wherein the proviso of set currents being zero is possible. According to the invention, each inverter is a self-commutated inverter with switchable power semiconductors and the rectifier regulation (7_rr) of the provided inverter (1) is regulated such that the sum of the product of the differential voltages (du) and the value of given set DC current ( \|Idco_r\|) at th corresponding rectifier and the differential current (di) is a minimum (du \| Idco_r \| +di -- >Min) and the inverter regulation (8_ii) regulates the corresponding inverter (1) such that the sum between the differential voltage (du) and the differential current (di) is a minimum (du+di -->Min).

Full Text

Description
Installation having a DC voltage intermediate circuit and self-
commutated converters
The invention relates to a method for closed-loop control of at
least two converters, which can be controlled as rectifiers or
inverters and are connected to one another via a DC voltage
link, in the field of power transmission and/or distribution,
in which a measurement DC voltage (Udc_r1...Udc_rr;
Udc_i1...Udc_ii) and a measurement direct current
(Idc_r1...Idc_rr; Idc_i1...Idc_ii) are in each case measured at
each converter and are transmitted to a rectifier regulator
for closed-loop control of the respective rectifier, or are
transmitted to an inverter regulator for closed-loop control of
the respectively associated inverter, wherein each rectifier
regulator and each inverter regulator in each case forms the
difference between a predetermined nominal DC voltage (Udco)
and the respectively received measurement DC voltage
(Udc_r1...Udc_rr; Udc_i1...Udc_ii), producing a difference DC
voltage (du) and, furthermore, the difference between a nominal
direct current (Idco_r1...Idco_rr, Idco_i1...Idco_ii) and the
respectively received measurement direct current
(Idc_r1...Idc_rr; Idc_i1...Idc_ii) , producing a difference
direct current (di), where the measurement direct current, the
measurement DC voltage, the nominal direct current and the
nominal DC voltage are in a normalized form.
A method such as this is already known from WO 2007/033620 Al,
which describes a closed-loop control method for high-voltage
direct-current transmission in which electrical power is
transmitted between AC voltage systems via a direct-current
circuit. The high-voltage direct-current transmission (HVDCT)
installation which is used

for power transmission consists of a rectifier and an inverter,
which are connected to one another via a direct current link.
Transformers are provided in order to couple the converters to
the respectively associated AC voltage system. For closed-loop
control of the rectifier or inverter, a measurement DC voltage
Udc_r or Udc_i, and a respective measurement direct current,
Idc_r or Idc_i respectively, are recorded and are transmitted
to the respective closed-loop control system both at the
rectifier and at the inverter. A nominal DC voltage Udco and a
nominal direct current Idco are determined from the
predetermined power which is intended to be transmitted, with
the aid of a function transmitter. The difference DC voltage du
is then calculated from the difference between the nominal DC
voltage Udco and the measurement DC voltage Udc_r, Udc_i. The
difference direct current di is correspondingly formed from the
difference between the nominal direct current Idco and the
respectively determined measurement direct current Idc_r and
Idc_i. All the values are in this case in normalized form, with
the normalization process being carried out, for example, with
respect to a rated DC voltage and a rated direct current, or
else with respect to a nominal direct current and the nominal
DC voltage. The rectifier regulator now controls the rectifier
in such a way that the sum of the difference DC voltage du and
of the difference direct current di is a minimum. The inverter
regulator in contrast controls the inverter such that the
difference between the difference direct current di and the
difference DC voltage du is a minimum. The already known
methods are, however, suitable only for so-called externally
commutated converters, in which, for example, thyristors are
used which cannot be turned off by means of trigger signals.
Converters such as these allow current to flow in only one
direction in the direct current circuit. The power flow can be
reversed only by reversing the polarity of the voltage across
the respective converter. Furthermore,

in this installation, which is referred to in the following
text as a traditional HVDCT installation, the operating current
must always be greater than 0.05 p.u., because of the need to
avoid intermittent currents in the current nominal value. This
allows the abovementioned normalization with respect to the
nominal values.
As a result of the progressive improvement in the field of
power electronics, it has become possible to also use power
semiconductors which can be turned off, for example IGBTs or
GTOs, for converters in the field of power transmission, and in
particular for HVDCT. A converter such as this, which is also
called a voltage source converter (VSC), is connected to a
further VSC via a DC voltage intermediate circuit. Each power
semiconductor which can be turned off in the converter has a
freewheeling diode connected in parallel with it. The power
flow is no longer reversed by reversing the polarity of the
voltage on the respective VSC but by reversing the current
flowing via the respective VSC. Furthermore, direct current
nominal values equal to zero can be used for the respective
VSC. It is therefore impossible to use the method of this
generic type for controlling a VSC in an HVDCT installation.
WO 2007/033619 A1 discloses a closed-loop control method for
direct current transmission having a plurality of converters,
in which the method mentioned above is likewise used. In this
case as well, nominal currents of more than 0.05 p.u. are
required, in particular at the rectifier, in order to avoid
intermittent currents.
The object of the invention is to provide a method of the type
mentioned initially, which allows closed-loop control to be
carried out for converters which have power semiconductors
which can be turned off and are connected to one another via a
DC voltage intermediate circuit.

while at the same time making it possible to preset nominal
currents equal to zero.
The invention achieves this object in that each converter is a
self-commutated converter (1) having power semiconductors which
can be turned off, and the rectifier regulator controls the
respectively associated converter such that the sum of the
product of the difference DC voltage (du) and the magnitude of
the nominal direct current (|Idco_r|), which is provided at the
respectively associated rectifier, and the difference direct
current (di) is a minimum (du*|Idco_r|+di—>Min) and the
inverter regulator controls the respectively associated
converter such that the sum between the difference DC voltage
(du) and the difference direct current (di) is a minimum
(du+di—>Min).
The invention transfers the closed-loop control method known
from traditional HVDCT to HVDCT installations in which VSCs and
a DC voltage intermediate circuit are used. In the case of
these so-called voltage source converters (VSC), the voltage
polarity can no longer be reversed because of the DC voltage
intermediate circuit, by which means the power flow is reversed
in a conventional HVDCT. In converters (VSCs) such as these,
the power flow is reversed by reversing the current flow. These
physical differences are taken into account in the closed-loop
control method according to the invention. Furthermore, the
method according to the invention also makes it possible to
preset nominal currents with a value of zero. This was
impossible in the case of the known closed-loop control methods
for conventional HVDCT. The method according to the invention
therefore provides simple and flexible closed-loop control for
two or more VSCs which are connected to one another via a DC
voltage link.

The DC voltage link has a polarity which does not change during
operation of the HVDCT installation, within the scope of the
invention, even when the power flow is reversed. In this case,
for example, the DC voltage link is a bipolar DC voltage link
which extends between two converters. According to this
variant, the HVDCT installation to be controlled consists of a
rectifier, which is connected via an inductance, for example a
transformer, to an AC voltage system which provides power. The
inverter, which is connected to said rectifier via the bipolar
DC voltage link, is likewise connected via an inductance, for
example a transformer, to a second AC voltage system, which,
for example, has the loads to be supplied. However, in contrast
to this, a plurality of converters can also be controlled using
the method according to the invention, wherein the converters,
or in this case the VSCs, are connected to one another via a DC
voltage link with any desired topology. The method according to
the invention is completely independent of the topology of the
DC voltage link.
Expediently, the nominal DC voltage Udco of each rectifier
regulator and the nominal DC voltage Udco of each inverter
regulator are identical. Complex definition of the nominal DC
voltage at each converter station, with rapid and reliable
telecommunication for signaling the respective nominal DC
voltage to other stations, has therefore become superfluous
within the scope of the invention.
Expediently, the converters are at a distance of at least
1 kilometer from one another. All HVDCT installations which use
the so-called back-to-back configuration, in which inverters
and converters are installed physically alongside one another
and are used only for coupling different AC voltage systems in
a manner which can be subjected to closed-loop control,

are excluded from this expedient further development. The
apparatus for which the method according to this advantageous
further development is designed is therefore used to transmit
electrical power over a relatively long distance.
Advantageously, each measurement DC voltage Udc_rr or Udc_ii
and the nominal DC voltage Udco are normalized with respect to
the nominal DC voltage Udco. This allows stable closed-loop
control, in particular when the power levels to be transmitted
are low. This normalization, which is already known from
conventional HVDCT, is, however, carried out for the purposes
of the exemplary embodiment according to the invention only in
conjunction with the respective measurement and nominal DC
voltage. In contrast, the nominal direct current is not
normalized, in order to allow the use of nominal direct
currents equal to zero in this expedient further development of
the invention as well.
Advantageously, two converters are provided, one of which is
operated as a rectifier and the other converter is operated as
an inverter, wherein a DC voltage intermediate circuit extends
between the rectifier and the inverter. According to this
refinement, the method is used for an HVDCT installation with
VSCs and a DC voltage intermediate circuit without any
branches. For example, the DC voltage intermediate circuit is
designed with one pole, using the ground as the return
conductor. In contrast to this, the DC voltage link has two
poles, with the positive pole and negative pole of the DC
voltage link being in the form of cable conductors.
According to one refinement, which differs from this, the
method according to the invention is provided for at least
three converters, between which a DC voltage system extends.

The DC voltage system may have any desired topology.
According to one expedient further development in this context,
a respectively associated rectifier nominal DC power
(Pdco_r1...Pdco_rr) or inverter nominal DC power
(Pdco_i1...Pdco_ii) is in each case defined for each rectifier
regulator and for each inverter regulator, respectively,
wherein the sum of all the rectifier nominal DC powers and all
the inverter nominal DC powers is equal to zero. By way of
example, the respective nominal DC power levels are defined by
a central control center and are transmitted from there to the
respective converter regulators. There is no need whatsoever
for a powerful communication link for transmission, for example
as is the case with a conventional HVDCT. For the purposes of
the invention, it is sufficient to transmit the respective
nominal DC power levels via the Internet, or any other simple,
and therefore cost-effective, form of telecommunication to the
respective regulator.
Expediently, the nominal direct current respectively associated
with the converter is determined at each converter from the
nominal DC power associated with it and from the nominal DC
voltage, which is common to all of them. According to this
advantageous further development, the method according to the
invention is simple and clear.
Expediently, the closed-loop control of each rectifier and each
inverter is carried out over the entire operating range of the
rectifier or of the inverter both on the basis of the
respectively associated rectifier difference direct current
(di_r1...di_rr) and on the basis of the rectifier difference DC
voltage (du_r1...du_rr) and, respectively, on the basis of the
respectively

associated inverter difference direct current (di_r1...di_rr)
and on the basis of the associated inverter difference DC
voltage (du_i1...du_ii).
Advantageously, a respective measurement DC voltage
(Udc_r1...Udc_rr; Udc_i1...Udc_ii) and a respective measurement
direct current (Idc_r1...Idc_rr; Idc_i1...Idc_ii) are measured
at each converter and are transmitted to a rectifier regulator
or to an inverter regulator.
According to one expedient further development in this context,
a number of converters are controlled such that a direct
current which flows through an isolating switch which is
arranged in the DC voltage link is equal to zero. According to
this advantageous further development, simple isolating
switches can be used in the DC voltage link, since switching
can always take place when no current is flowing. Complex
resonant circuits which apply a zero crossing to the current by
means of the switch, thus extinguishing any arc which is struck
on switching, have likewise been rendered superfluous by this
advantageous refinement of the invention.
According to one modification in this context, at least one
converter controls the respective measurement direct current
recorded at it at zero, and at least one isolating switch,
which is arranged in the DC voltage link, is then opened. Zero
current regulation is carried out by using a nominal direct
current equal to zero. According to one preferred refinement of
this further development, all the converters which are
connected to one another by a DC voltage link are controlled to
zero. In other words, a nominal direct current zero is used for
each regulator.

According to one expedient further development in this context,
a switching-off time is set and is transmitted to the
respectively affected regulators, wherein said regulators
control the measurement direct current associated with them to
be zero when the switching-off time is reached. According to
this expedient further development, all the regulator units
have a time signal transmitter, for example an accurate clock,
which provides a common, or substantially common, time signal
for all the regulators. This time signal is compared with the
set and transmitted switching-off time. If the difference
between the switching-off time and the measured time is less
than a threshold value set in advance, that is to say when the
switching-off time is reached, the respective regulating unit
presets a nominal direct current of zero, such that the
measurement direct current is controlled at zero. This
refinement ensures that all the affected regulators which, for
example, are selected as such by a central control station, or
all the regulators, are transferred to zero-current control at
the same time.
According to one expedient further development in this context,
the measurement direct current is controlled to be zero by each
affected converter over a zero-current time period, and the
normal closed-loop control method is then restarted, that is to
say once the zero-current time period has elapsed. Within the
zero-current time period which, for example, is in the region
of a few milliseconds, the desired isolating switch or switches
in the DC voltage link is or are now opened. This makes it
possible, in a simple manner, to deliberately disconnect an
area or a converter from the DC voltage system with any desired
topology, without likewise having to interrupt the power
transmission via those links or converters which have not been
turned off. In fact, within the scope of the invention, it is
possible

to disconnect a section of the DC voltage link or a converter
specifically from the DC voltage system by rapidly reducing the
direct current. The normal closed-loop control of the HVDCT
installation is then started again. The closed-loop control
method according to the invention then automatically approaches
the regulator operating points required for this purpose,
without any knowledge of the new topology of the DC voltage
link. Complex data transmissions, calculations or the like have
become superfluous as a result of the invention.
Further expedient refinements and advantages of the invention
are the subject matter of the following description of
exemplary embodiments of the invention, with reference to the
figures of the drawing, in which the same reference symbols
refer to components having the same effect, and in which:
Figure 1 shows an HVDCT installation having a plurality
of voltage source converters, which are
connected to one another via a DC voltage
system,
Figure 2 shows parameters and controlled variables used
by one exemplary embodiment of the closed-loop
control method according to the invention,
Figure 3 shows one exemplary embodiment of the method
according to the invention for a plurality of
VSCs, which are linked to one another via a DC
voltage system, and
Figure 4 shows one exemplary embodiment of the method
according to the invention, in which one
specific section of the DC voltage link

is disconnected from the DC voltage system by
means of a single isolating switch.
Figure 1 schematically illustrates an HVDCT installation having
a multiplicity of converters 1, which are each connected via a
transformer 2 to an AC voltage system 3. Each converter 1 is a
so-called voltage source converter (VSC), having power
semiconductors which can be turned off, such as IGBTs or GTOs,
each of which has a freewheeling diode connected back-to-back
in parallel with it. In this case, each converter 1 is
connected to a DC voltage system 4 as a DC voltage link, which
DC voltage system 4 may have any desired topology. Furthermore,
each VSC 1 has an associated smoothing inductor 5. A VSC 1, a
transformer 2, and a smoothing inductor 5 are part of a
rectifier station or inverter station, depending on how the
respective VSC is controlled by the regulating unit or
regulator, which is not illustrated in the figure. In Figure 1,
those rectifier stations which have VSCs operated at rectifiers
are annotated r1, r2, r3 ... rr, whereas those inverter
stations with VSCs which are operated as inverters are
annotated i1, i2, i3 ... ii. Figure 1 also shows that each
converter station rr and each inverter station ii has sensors
for recording the measurement DC voltage Udc_rr at this station
as well as the direct current Idc_rr or Idc_ii flowing there at
the VSC 1. As can likewise been seen from Figure 1, the
respective AC voltage systems 3 are also connected to one
another via AC voltage links 6. The alternating current link 6
does not adversely affect the method according to the invention
for closed-loop control of the VSC.
Figure 2 is a graph, in which the normalized direct current is
shown on the abscissa and the normalized DC voltage is shown on

the ordinate. This is based on the assumption that, in contrast
to the illustration in Figure 1, only one rectifier is
connected via a DC voltage intermediate circuit to a VSC
operating as an inverter.
In the graph, the parameters and controlled variables of the
rectifier are shown in the left-hand quadrant of the graph, and
corresponding variables for the inverter are shown in the
right-hand quadrant. The measurement direct current and the
measurement DC voltage at the measurement point X_r are plotted
on the rectifier side, whose measured and nominal values are
annotated _r. The measurement DC voltage Udc_r is therefore
greater than the nominal DC voltage Udco or Udco_r. If the
difference DC voltage du_r is formed and normalized with
respect to the nominal DC voltage Udco, this results in
. This value must be less than zero for power
to
be transmitted to the inverter.
The nominal direct current Idco_r of the rectifier is, by-
definition, negative, as a result of which: Idco_r=-Idco. The
difference direct current di, which is formed from the
difference between the nominal and the measurement direct
current, is then given by: di_r=-Idco_r+Idc_r. This is not
normalized with respect to the nominal direct current Idco_r,
in order to allow a nominal direct current of zero for closed-
loop control. The closed-loop control is now carried out by
minimizing the rectifier control error de_r=du_r*|Idco_r|+di_r
being, that is to say becoming zero. The straight line shown on
the left is therefore a tangent to the hyperbola which is shown
by a dashed line and represents the invariant nominal DC power
Pdco_r=-Pdco. Since the difference DC voltage du_r is
multiplied by the magnitude of the nominal direct current
|Idco_r| in order to calculate the control error de_r, this

results in a stable, rapidly active control response of the
rectifier, because the closed-loop voltage control is
suppressed when the nominal direct currents are small. Closed-
loop voltage and current control are carried out,
simultaneously and in a substantially equivalent form to one
another, only for higher nominal direct currents Idco_r.
The measurement direct current Idc_i and the measurement DC
voltage Udc_i at the operating point X_i are plotted on the
inverter side. As on the rectifier side, the difference DC
voltage du_i and the difference direct current di_i are
calculated, with the difference voltage du_i once again being
normalized with respect to the nominal DC voltage Udco. du_i
must be greater than zero for the desired power transmission.
Normalization with respect to the nominal direct current in
order to form the difference direct current di_i is also not
carried out in this case. It is therefore also possible to
preset nominal DC currents Idco_i equal to 0. For the desired
power transmission, the difference direct current di_i is less
than zero. The control error de_i is formed from the sum of the
difference DC voltage du_i and the difference direct current
di_i. The closed-loop control now attempts to minimize the
control error de_i to be zero. On the graph shown in Figure 2,
this results in a straight line which intersects the hyperbola
of the invariant DC power Pdco_i at the point W_i. The straight
line de_i is reminiscent of the response of a pure resistance,
as a result of which the inverter regulator can also be
referred to as a resistance regulator.
Figure 3 schematically illustrates one exemplary embodiment of
the method according to the invention for an HVDCT installation
as shown in Figure 1, in which the parameters and variables
shown in Figure 2 are used for a rectifier regulator 7_r1 for
the rectifier station r_1, and for an inverter regulator 8_i1
for the inverter station i1. The rectifier regulators 7_rr,

which are not shown in the figure, and the inverter regulator
8_ii, which is not illustrated, are physically identical.
As can be seen from Figure 3, the rectifier regulator 7_r1, and
therefore each rectifier regulator, receives a nominal DC power
Pdco_r1 assigned to it. A corresponding situation applies to
the inverter regulator 8_i1, in which case the respectively
associated nominal DC power is transmitted from a central
control station to the respective stations via communication
links which are not shown, for example a simple radio link. In
this case, the sum of all the nominal DC powers is equal to
zero: ?Pdco_rr+?Pdco_ii=0. The invention makes complex, rapid
and reliable transmission, as is the case for conventional
HVDCT, superfluous.
The received nominal DC power Pdco_r1 and Pdco_i1 is in each
case supplied to a divisor 9, to whose second input the nominal
DC voltage Udco, which is the same for all the stations, is
applied. By way of example, the nominal DC power is likewise
transmitted from a central station. The divider 9 forms the
quotient of the respective nominal DC power Pdco and the
nominal DC voltage Udco, producing the nominal direct current
Idco_r1 or Idco_i1, with the respective nominal direct current
being supplied to a limiter 10, which limits the nominal direct
current Idco_r1 or Idco_i1 to a respective minimum nominal
direct current Imin_r1,Imin_i1 and a respective maximum
nominal direct current Imax_r1,Imax_i1. The nominal direct
current Idco_r1 or Idco_i1 is then supplied to an adder 11,
which in each case forms the difference from the nominal direct
current Idco_r1 or Idco_i1 and the measurement direct current
Idco_r1 or Idco_i1. The difference direct current di_r1
obtained in this way is then supplied to a further adder 11.
The value at the second

input of said adder 11 is derived from the measured voltage and
the nominal voltage. For this purpose, the divisor 12 in each
case determines the quotient of the measurement DC voltage
Udc_r1 or Udc_i1 and the nominal DC voltage Udco. The
measurement DC voltage Udc_r1 or Udc_i1 which have been
normalized in this way is then subtracted from 1, producing the
respective difference DC voltage du_r1 or du_i1. As stated
above, for the inverter regulator 8_il, the difference DC
voltage du_i1 obtained in this way is supplied to the second
input of the adder 11 which, by addition of its inputs,
calculates the inverter control error de_i, which is then
supplied to a regulator 13.
In contrast to the inverter regulator 8_i1, in the case of the
rectifier regulator 7_r1,the difference DC voltage du_r1 is
multiplied by the magnitude of the nominal direct current
|Idco_r1|. An absolute-value generator 23 is provided in order
to form the magnitude |Idco_r1| from Idco_r1,with the product
|Idco_r1| * du_r1 being formed by means of a multiplier 24. In
the case of the rectifier regulator 7_r1,the difference direct
current di_r1 is added to said product of du_r1 and |Idco_r1|
by means of the adder 11, in order to produce the control error
de_r of the rectifier. The control error of the rectifier de_r1
is then supplied to the regulator 13 and, finally, to a module
management system 14, which in this case provides open-loop
control for the power semiconductors in the respective VSCs.
The VSCs illustrated in Figure 3 are so-called multilevel VSCs
which, like all VSCs, consist of a bridge circuit of electrical
valves. However, in the case of multilevel VSCs, each
electrical valve is formed from a series circuit of bipolar
submodules which each have an energy store and, in parallel
with the energy store, a circuit composed of

power semiconductors, such that the voltage across the energy
store, or else a zero voltage, is dropped across the respective
submodule depending on the operation of the power
semiconductors. The voltage which is dropped in total across
the electrical valves is additively composed of the output
voltages of the submodules, and can therefore be changed in
steps.
In order to operate the power semiconductors in the submodules,
the output of the regulator 13 is connected to the input of the
so-called module management system, whose detailed
configuration will not be described for the purposes of the
invention, since it is not essential to the invention. The
module management system operates the power semiconductors in
the submodules corresponding to the output of the regulator 13.
However, within the scope of the invention, it is also possible
for the regulator 13 to be followed by a pulse width modulator
rather than by a module management system, which pulse width
modulator is designed for open-loop control of two-stage or
three-stage voltage source converters.
Figure 4 likewise shows an HVDCT installation, in order to
illustrate one exemplary embodiment of the method according to
the invention. In this case as well, a plurality of rectifier
stations r1, r2,...rr and a plurality of inverter stations i1,
i2...ii are once again connected to one another via a DC
voltage system 4 of any desired topology. In addition to the
illustration shown in Figures 3 and 1, each rectifier station
rr and each inverter station ii has a protection unit 15 in
addition to a rectifier regulator 7_rr or inverter regulator
8_ii, respectively, which protection unit 15 is connected to
the respective rectifier regulator 7_rr or inverter regulator
8_ii. The respective rectifier regulator 7_rr or inverter
regulator 8_ii is also in each case connected to a protection
device 16 of a circuit breaker 17, with the switch 17 being
arranged between the AC

voltage system 3 and the transformer 2 . In the event of a
fault, for example high short-circuit currents, it is possible
to disconnect each VSC from the respective AC system 3 by means
of the multi-pole switch 17 on the AC voltage side which, for
example, is a circuit breaker for switching high short-circuit
currents. In this case, protection tripping takes place via the
protection unit 15, which causes the respective regulating unit
7 to emit a tripping command to the protection device 16. The
protection device 16 opens the circuit breaker 17 when a
tripping command occurs.
In order to allow DC voltage sections 18 to also be turned off
or disconnected within the DC voltage link 4, DC voltages
switches 19 are arranged in the DC voltage links. A DC voltage
protection unit 2 0 is used to trip the DC voltage switch or
switches 19 and is connected to the output of a DC voltage
current sensor 21, which produces direct current values Idc_bl
or Idc_b2, corresponding to the direct current flow via the
respective switch 19. The DC voltage protection unit 2 0 is in
turn connected to the protection device 16 for the DC voltage
switch 19.
The arrow 22 shown in Figure 4 indicates a ground fault, as a
result of which high short-circuit currents flow in the DC
voltage section 18. If the direct current values Idc_bl and
Idc_b2 which are supplied to the respective DC voltage
protection unit 2 0 exceed a previously defined threshold value,
or some other criterion, the respective DC voltage protection
unit 20 defines a switching-off time toff in the near future,
and transmits the switching-of f time toff to the respective
rectifier regulator 7_r1,7_r2, 7_rr or inverter regulator
8_i1, 8_i2, 8_ii. These regulators are connected to a timer,
such that

all the regulators can determine that the switching-off time
has been reached, approximately at the same time. The
protection unit 15 now operates the respective regulating unit
such that this in each case sets the respective nominal direct
current Idco_r1,Idco_r2, Idco_rr or Idco_i1, Idco_i2 and
Idco_ii to zero throughout a zero-current time period, such
that the respective measurement direct current is controlled at
zero. Therefore, the respective direct current Idc_b1 or Idc_b2
flowing through the direct current switches 19 is also equal to
zero. The switch 19 can now be opened with no current flowing.
The DC voltage protection unit 20 likewise identifies that the
switching-off time has been reached, by time comparison. After
a safety time interval has passed, which is shorter than a
zero-current time period, the DC voltage switch 19 is opened
with no current flowing, thus disconnecting the DC voltage
section 18. The normal closed-loop control method is started
again once the zero-current time period has elapsed. For the
purposes of the invention, the respective regulators need not
be informed of the change in the topology of the DC voltage
link 4. The closed-loop control system automatically changes to
stable closed-loop control points without any further
additional effects. This allows the DC voltage system 4 to be
switched with little complexity. The invention renders parallel
resonant circuits, as used for DC voltage switches according to
the prior art, superfluous.

WE CLAIM
1. A method for closed-loop control of at least two
converters (1), which can be controlled as rectifiers or
inverters and are connected to one another via a DC
voltage link (4), in the field of power transmission
and/or distribution, in which
a measurement DC voltage (Udc_r1...Udc_rr;
Udc_i1...Udc_ii) and a measurement direct current
(Idc_r1...Idc_rr; Idc_i1...Idc_ii) are in each case
measured at each converter (1) and are transmitted to a
rectifier regulator (7_rr) for closed-loop control of
the respective rectifier, or are transmitted to an
inverter regulator (8_ii) for closed-loop control of
the respectively associated inverter,
wherein each rectifier regulator (7_rr) and each
inverter regulator (8_ii) in each case forms the
difference between a predetermined nominal DC voltage
(Udco) and the respectively received measurement DC
voltage (Udc_r1...Udc_rr; Udc_.i1...Udc_ii), producing a
difference DC voltage (du) and, furthermore, the
difference between a nominal direct current
(Idco_r1...Idco_rr, Idco_i1...Idco_ii) and the
respectively received measurement direct current
(Idc_r1...Idc_rr; Idc_i1... Idc_ii), producing a
difference direct current (di), where the measurement
direct current, the measurement DC voltage, the nominal
direct current and the nominal DC voltage are in a
normalized form,
characterized in that
- each converter is a self-commutated converter (1)
having power semiconductors which can be turned off,
and

the rectifier regulator (7_rr) controls the
respectively associated converter (1) such that the sum
of the product of the difference DC voltage (du) and
the magnitude of the nominal direct current

(|Idco_r|) , which is provided at the respectively-
associated rectifier, and the difference direct current
(di) is a minimum (du*|Idco_r|+di?Min) and
the inverter regulator (B_ii) controls the respectively-
associated converter (1) such that the sum between the
difference DC voltage (du) and the difference direct
current (di) is a minimum (du+di?Min).
2. The method as claimed in claim 1,
characterized in that
the nominal DC voltage (Udco) of each rectifier regulator
and the nominal DC voltage (Udco) of each inverter
regulator are identical.
3. The method as claimed in claim 1 or 2,
characterized in that
the converters (1) are installed at a distance of at least
1 km from one another.
4. The method as claimed in one of the preceding claims,
characterized in that
each measurement DC voltage (Udc_rr) and the nominal DC
voltage (Udco) are normalized with respect to the nominal
DC voltage (Udco).
5. The method as claimed in one of the preceding claims,
characterized in that
two converters (1) are provided, one of which is operated
as a rectifier and the other converter (1) is operated as
an inverter, wherein a DC voltage intermediate circuit (4)
extends between the rectifier and the inverter.
6. The method as claimed in one of claims 1 to 4,
characterized in that
at least three converters (1) are provided, between which
a DC voltage system (4) extends.

7. The method as claimed in claim 6,
characterized in that
a respectively associated rectifier nominal DC power
(Pdco_r1...Pdco_rr) or inverter nominal DC power
(Pdco_i1...Pdco_ii) is defined for each rectifier
regulator (7_rr) and for each inverter regulator (8_ii),
respectively, wherein the sum of all the rectifier nominal
DC powers and all the inverter nominal DC powers
(?Pco_rr+?Pdco_ii) is equal to zero.
8. The method as claimed in claim 7,
characterized in that
the nominal direct current (Idco_rr; Idco_ii) associated
with the respective converter (1) is determined at each
converter (1) from the nominal DC power (Pdco_rr; Pdco_ii)
associated with it and from the nominal DC voltage (Udco),
which is common to all of them.
9. The method as claimed in one of the preceding claims,
characterized in that
the closed-loop control of each rectifier and each
inverter is carried out over the entire operating range of
the rectifier or of the inverter both on the basis of the
respectively associated rectifier difference direct
current (di_r1...di_rr) and on the basis of the rectifier
difference DC voltage (du_r1...du_rr) and, respectively,
on the basis of the respectively associated inverter
difference direct current (di_rl...di_rr) and on the basis
of the associated inverter difference DC voltage
(du_i1...du_ii).

10. The method as claimed in one of the preceding claims,
characterized in that
a respective measurement DC voltage (Udc_r1...Udc_rr;
Udc_i1...Udc_ii) and a respective measurement direct
current (Idc_rl. . . Idc_rr; Idc_.i1...Idc_ii) are measured at
each converter (1) and are transmitted to a rectifier
regulator (7_rr) or to an inverter regulator (8_ii).
11. The method as claimed in claim 10,
characterized in that
a number of converters (1) are controlled such that a
direct current (Idc_b1;Idc_b2) which flows through an
isolating switch which is arranged in the DC voltage link
(4) is equal to zero.
12. The method as claimed in claim 10 or 11,
characterized in that
at least one converter (1) controls the respective
measurement direct current (Idc_rr;Idc_ii) recorded at it
at zero, and at least one isolating switch (19), which is
arranged in the DC voltage link (4), is then opened.
13. The method as claimed in claim 12,
characterized in that
a switching-off time (toff) is set and is transmitted to
the respectively affected regulators, wherein said
regulators control the measurement direct current
(Idc_rr; Idc_ii) associated with them to be zero when the
switching-off time (toff) is reached.
14. The method as claimed in one of claims 11 to 13,
characterized in that

the measurement direct current (Idc_rr;Idc_ii) is
controlled to be zero by each affected converter (1) over
a zero-current time

period, and the normal closed-loop control method is then
started.
15. The method as claimed in one of claims 11 to 14,
characterized in that
one of the converters (1) is disconnected from the DC
voltage link (4) by opening the isolating switches (19).
16. The method as claimed in one of claims 11 to 15,
characterized in that
a section (18) of the DC voltage link (4) is selectively-
disconnected from the DC voltage link (4) by switching the
isolating switches (19).

The invention relates to a method for regulating at least two inverters (1),
which may be regulated either as rectifier or inverter, connected to each
other by a DC link (4), for energy transmission and/or distribution, wherein
a measured DC voltage (Udc_rl... Udc_rr; Udc_il... Udc_ii) and a measured
DC current ( Idc_rl... Idc_rr; Idc_il... Idc_ii ) is measured at each inverter
(1) and transmitted to a rectifier regulator (7_rr) for regulating the
corresponding rectifier or to an inverter regulator (8_ii) for regulating the
corresponding inverter, wherein each rectifier regulator (7_rr) and each
inverter regulator (8_ii) give the difference between a given set DC voltage
(Udco) and the relevant received measured DC voltage (Udc_rl... Udc_rr,
Udc_il... Udc_ii) to give a differential DC voltage (du) and the difference
between a set DC current (Idco_rl Idco_rr, Id-co_il Idco_n) and the
corresponding received measured DC current (Idc_rl... Idc_rr, Idc_il...
Idc_ii) to give a differential DC current (di), wherein the measured DC
current, the measured DC voltage, the set DC current and the set DC voltage
are normalised with which a regulation of inverters comprising switchable
power semiconductors connected by a DC link can be carried out, wherein
the proviso of set currents being zero is possible. According to the invention,
each inverter is a self-commutated inverter with switchable power
semiconductors and the rectifier regulation (7_rr) of the provided inverter
(1) is regulated such that the sum of the product of the differential voltages
(du) and the value of given set DC current ( |Idco_r|) at th corresponding
rectifier and the differential current (di) is a minimum (du* | Idco_r | +di --
>Min) and the inverter regulation (8_ii) regulates the corresponding inverter

(1) such that the sum between the differential voltage (du) and the
differential current (di) is a minimum (du+di -->Min).

Documents:

http://ipindiaonline.gov.in/patentsearch/GrantedSearch/viewdoc.aspx?id=lTAPlslmmST6+fqLzn2eXg==&loc=wDBSZCsAt7zoiVrqcFJsRw==

« Previous Patent

Next Patent »

Patent Number

278705

Indian Patent Application Number

4754/KOLNP/2010

PG Journal Number

54/2016

Publication Date

30-Dec-2016

Grant Date

28-Dec-2016

Date of Filing

13-Dec-2010

Name of Patentee

SIEMENS AKTIENGESELLSCHAFT

Applicant Address

WITTELSBACHERPLATZ 2 80333 MÜNCHEN GERMANY

Inventors:

#	Inventor's Name	Inventor's Address
1	EULER, INGO	SPITZWEGSTR. 6, 91056 ERLANGEN, GERMANY
2	WÜRFLINGER, KLAUS	POPPENREUTHER STR. 49A, 90419 NÜRNBERG, GERMANY
3	STRAUSS, JOHN-WILLIAM	15 RED HILL CIRCLE APT-L, 94920 TIBURON 94920, UNITED STATES OF AMERICA
4	TU, QUOC-BUU	SICHERSDORFER STR. 28, 90574 ROSSTAL, GERMANY
5	WITTSTOCK, CARSTEN	BEIM GRÖNACKER 21, 90480 NÜRNBERG, GERMANY
6	RITTIGER, JÜRGEN	KARL-BRÖGER-STR. 39, 91074 HERZOGENAURACH, GERMANY
7	VENJAKOB, OLIVER	LEUSCHNERWEG 10, 91058 ERLANGEN, GERMANY
8	BERNHARD, TOBIAS	JULIUS-RING 46, 96114 HIRSCHAID, GERMANY
9	DOMMASCHK, MIKE	HEINRICH-BOLZE-STR. 10, 03044 COTTBUS, GERMANY
10	DORN, JÖRG	MOZARTSTR. 19, 96155 BUTTENHEIM, GERMANY
11	KARLECIK-MAIER, FRANZ	TILMANN-RIEMENSCHNEIDER-STR. 63, 91315 HÖCHSTADT, GERMANY
12	LANG, JÖRG	KRONACHER STRASSE 14, 95346 STADTSTEINACH, GERMANY

PCT International Classification Number

H02J 3/36

PCT International Application Number

PCT/EP2008/005211

PCT International Filing date

2008-06-17

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1			NA