Method for controlling a self-commutated power converter, self-commutated power converter and system comprising the self-commutated power converter

Abstract

Method for controlling a self-commutated converter (202, 316, 402), wherein at least one current electrical parameter of a system (302) is provided, wherein: - at least one current time parameter of the system (302) is determined from the current electrical parameter, - at least one previous pointer control signal is provided, wherein the previous pointer control signal was determined in a previous control interval, - the previous pointer control signal is rotated by an angle, wherein the angle is determined as a function of the determined current time parameter, and - the converter (202, 316, 402) is controlled as a function of the rotated pointer control signal, - wherein a current pointer control signal is determined from the rotated pointer control signal, wherein the current pointer control signal is determined at least as a function of the rotated pointer control signal and at least one setpoint parameter to realize a specific system characteristic.- wherein the at least one setpoint parameter is a setpoint apparent power,- wherein a first setpoint indicator signal is determined from the rotated indicator position signal and the setpoint apparent power,- wherein a second setpoint indicator signal is determined from the current of the line (302) and at least the first setpoint indicator signal by a maximum value correction, characterized in that- a previous first setpoint indicator signal or a previous second setpoint indicator signal is provided,- the previous first setpoint indicator signal and / or the previous second setpoint indicator signal is rotated by an angle, wherein the angle is determined as a function of the determined current time parameter, and- the rotated setpoint indicator signal and the first setpoint indicator signal are summed, and / or- the rotated setpoint indicator signal and the second setpoint indicator signal are summed.

Description

The invention relates to a method for controlling a self-commutated power converter, wherein at least one current electrical parameter of a system is provided. The invention further relates to a self-commutated power converter and a system comprising at least one self-commutated power converter. Today, conventional power plants, such as coal-fired, gas-fired, and nuclear power plants, are increasingly being replaced by renewable energy plants. Examples of renewable energy plants include wind turbines, photovoltaic systems, hydroelectric power plants, and biomass plants. It is estimated that by the end of 2020, renewable energy plants with an installed capacity of approximately 100,000 MW will be connected to the German electricity grid. However, a higher proportion of renewable energy generation plants and the associated displacement of conventional power plants leads to problems with grid stability. This is due to the grid-stabilizing function of the increasingly displaced conventional power plants and the grids and consumers that are adapted to this function, as well as the lack of grid-stabilizing capabilities in renewable energy generation plants. The operation of a conventional power plant is based on a three-phase AC power grid. The behavior of an AC power grid can be described by assuming a voltage source with symmetrical, equal source impedance. In this case, the magnitudes of the source voltages are equal, and an angle of 120° is maintained between them. Conventional power generation plants are configured to supply currents in both positive-sequence and negative-sequence modes. Through this operating principle, conventional power plants provide electrical energy by supplying their symmetrical AC power system, where the injected currents counteract the voltage symmetry. Depending on changes in load and grid conditions, the magnitude and angle of the source voltage also change rapidly, particularly instantaneously, thus stabilizing the grid.In other words, the grid is sufficiently stabilized by the conventional power generation plants themselves. Unlike conventional power plants, renewable energy plants are increasingly operated not according to the operating principle of the three-phase alternating current system, but rather according to the operating principle of the three-phase alternating current system. A corresponding three-phase alternating current system is characterized by the fact that the energy generated in the power plants is converted by power converters, in particular full converters. The converted energy can then be fed into the grid. In other words, the power converters imprint a current into the grid. In this context, self-commutated converters can be used in particular, which can measure the input variables required for their control directly from the grid into which the electricity is to be fed. The higher the proportion of converters in a grid section, the lower the short-circuit power available in that grid section relative to the individual generating units. Current methods for controlling power converters typically involve filtering rotary transformations, particularly dq-based approaches, where three-phase quantities with axes U, V, W can be transformed into a two-axis coordinate system with axes d and q. However, such approaches only allow for relatively slow control of a power converter and therefore, in the context of the significantly increased demands of modern grid structures, do not provide sufficient grid stabilization. This already leads to problems in grid sections dominated by power converters, limiting the operation of the corresponding renewable energy generation plants. This problem will be significantly exacerbated by the increasing share of renewable energy generation plants.In the future, power grids will therefore no longer be operable as usual, since the share of grid-stabilizing conventional power generation plants will be too low. In particular, power-intensive industrial plants will no longer be able to be supplied with sufficient short-circuit power from the public grid. In any case, the grid stabilization capabilities of conventional power generation plants—especially in terms of response speed—are insufficient to guarantee secure grid operation in the future. In this respect, they are even more likely to lead to grid destabilization. To ensure sufficient grid stability in the future, methods are needed that enable faster control of power converters. With real-time control methods known from the prior art, an electrical parameter of the system, in particular the current voltage and / or current of the system, can first be determined. The system can include the load and / or the grid. According to this method, a specific time-dependent voltage can be converted into a frequency-dependent voltage signal using network analysis. The frequency-dependent voltage signal can then be derived from the time-dependent voltage signal, for example, using a Discrete Fourier Transform (DFT) and / or a Phase-Locked Loop (PLL) algorithm. According to the state of the art, the frequency, phase, and amplitude of the fundamental frequency are determined in particular. Furthermore, a desired system characteristic, to be implemented at the converter's terminal, can be specified. From the frequency-dependent voltage signal and the specified system characteristic, a frequency-dependent target current, in particular the grid current to be realized, can then be generated. Error correction can also be performed. From the filtered slow pointer values, especially the frequency-dependent target current, (instantaneous) time-dependent control signals can be generated, for example, by feedforward control and a reference variable model. For instance, a time-dependent control voltage can be determined from the frequency-dependent target current by feedforward control based on an identified or estimated model of the system. This voltage can then be fed to a pulse generator to control a pulse inverter, i.e., the power converter. A disadvantage of this method is its high complexity. Time-dependent electrical parameters are first transformed into the frequency domain by filtering elements. The electrical signal is then further processed in the frequency domain. Subsequently, a suitable time-dependent control signal is generated based on the frequency-dependent signal. This method requires, in particular, filtering elements as well as a current controller. The publication by Aurtenechea, S.; Rodriguez, MA; Oyarbide, E.; Torrealday, JR; "Predictive Direct Power Control - A New Control Strategy for DC / AC Converters" IEEE Industrial Electronics, IECON 2006 - 32nd Annual Conference on Publication Year: 2006, Page(s): 1661 - 1666, discloses a method for controlling a self-commutated power converter, wherein at least one current electrical parameter of a system is provided. It is therefore an object of the present invention to provide a method for controlling a power converter which enables timely, in particular instantaneous, control of a power converter in a simple manner. The previously derived and described problem is solved according to the invention by a method according to claim 1 for controlling a self-commutated converter, wherein at least one current electrical parameter of a system is provided. At least one current time parameter of the system is determined from the current electrical parameter. At least one previous pointer position signal is provided, wherein the previous pointer position signal was determined in a previous positioning interval. The previous pointer position signal is rotated by an angle, wherein the angle is determined as a function of the determined current time parameter. The converter is controlled as a function of the rotated pointer position signal.A current pointer signal is determined from the rotated pointer signal, whereby the current pointer signal is determined at least as a function of the rotated pointer signal and at least one setpoint parameter to realize a specific system characteristic. The at least one setpoint parameter is a setpoint apparent power. A first setpoint pointer signal is determined from the rotated pointer signal and the setpoint apparent power. A second setpoint pointer signal is determined from the current of the system and at least the first setpoint pointer signal by means of a maximum value correction. A previous first setpoint pointer signal or a previous second setpoint pointer signal is provided. The previous first setpoint pointer signal and / or the previous second setpoint pointer signal is rotated by an angle, whereby the angle is determined as a function of the calculated current time parameter.The rotated setpoint signal and the first setpoint signal are summed and / or the rotated setpoint signal and the second setpoint signal are summed. In contrast to the prior art, according to the invention, by rotating a previous control signal by an angle depending on a current time signal, a timely, in particular instantaneous, control of a self-commutated power converter is possible in a simple manner. In a first step, a currently present electrical signal or an currently present electrical parameter of the system can be determined. This parameter is, in particular, an instantaneous, time-dependent parameter. According to a preferred embodiment of the method according to the invention, a current voltage of the system can be determined, in particular measured. Alternatively or additionally, the current of the system can be recorded. The system, or control system, can, in particular, comprise a load and a network. In contrast to the prior art, at least one current electrical parameter is used to derive (only) at least one current time parameter of the system. This at least one current time parameter can, in particular, be one or more parameters that allow conclusions to be drawn about the current phase angle of an electrical signal of the system. Preferably, it can be used to determine where the phase angle 0 is currently located. Depending on at least one current time parameter, a previous pointer control signal can be advanced by an angle. A previous pointer control signal is understood to be a control signal that was used to control the power converter in a previous control interval, particularly in the immediately preceding control interval. According to the invention, it can be assumed that the previous pointer setting signal was correct. It has been found, in particular, that a current setting signal can be determined from the previous setting signal by ascertaining the angle by which the previous setting signal has rotated. To determine the angle, at least one previously described current time parameter is determined. By rotating the previous control signal by this angle, the rotated pointer control signal can be rotated in such a way that, assuming an ideal system, the rotated pointer control signal corresponds to the pointer control signal currently to be realized. The power converter to be controlled can be regulated depending on the pre-rotated pointer control signal. It should be noted that this is generally a multi-phase system. Parameters or signals can therefore preferably be represented by space vectors. In particular, a vector signal can be an instantaneously defined or measured space vector. The method according to the invention eliminates the need for filtering elements and a current regulator. A transformation to the frequency domain is also unnecessary. Preferably, the method for controlling a power converter can be carried out (exclusively) in the time domain. In particular, sufficient grid stability can be ensured even in grid areas with a particularly high proportion of renewable energy generation plants. A high level of supply security can be guaranteed. According to a first embodiment of the method according to the invention, the at least one current time parameter can be a current mains frequency of the system. Alternatively or additionally, the at least one current time parameter can be a current sampling time of the system. The determined angle for rotating the previous pointer control signal can essentially correspond to the angle by which the previous pointer control signal ideally rotated from the previous control time to the current control time with the current mains frequency and / or the current sampling time, i.e., in an ideal rotating network, such as a three-phase power network. It can be assumed that this is an ideal system. In particular, a previous pointer control signal, such as a voltage space pointer or a current space pointer, can be pre-rotated by an angle by which a three-phase power system with the determined current mains frequency would have changed within the current sampling time.Error correction can be carried out in subsequent steps, as can the realization of a predefined system characteristic at the terminal of a current controller, such as a constant resistance or a constant power. As previously explained, the current voltage of the system can be determined as a current electrical parameter. Alternatively or additionally, the current current of the system can be determined as a current electrical parameter. From these time-dependent parameters, at least one current time parameter of the system, such as a mains frequency and / or a sampling time, can be easily determined. Furthermore, an electrical pointer control signal, preferably a voltage pointer control signal, can be defined as the pointer control signal. A voltage pointer control signal is particularly suitable for controlling a power converter. For example, a pulse width modulator can be provided that can convert the voltage pointer control signal into a suitable signal for controlling a pulse inverter. It is understood that, in the case of a voltage pointer setting signal, at least the current voltage of the line can be determined as the current electrical parameter, and corresponding time parameters for determining the angle can be derived from it. By controlling a power converter, a specific system characteristic can be implemented at the connection point (at any given time). Preferably, the behavior of a constant resistance can be emulated when energy is to be drawn from the grid, and a constant power output at the connection point can be implemented when energy is to be fed into the grid. The pointer control signal can be determined based on a predefined system characteristic. According to a preferred embodiment of the method according to the invention, a current pointer position signal is determined from the rotated pointer position signal. The current pointer position signal is determined at least as a function of the rotated pointer position signal and at least one setpoint parameter to realize a specific system characteristic. A specific system characteristic can thus be easily realized from the rotated pointer position signal and one or more setpoint values. In principle, various target parameters can be specified to achieve a specific system characteristic. According to a preferred embodiment of the method of the present invention, at least one of the target parameters is a target apparent power. A first target pointer signal is determined from the rotated pointer position signal and the target apparent power. Furthermore, a second target pointer signal is determined from the current of the system and at least the first target pointer signal by means of a maximum value correction. The target apparent power, i.e., a target active power and a target reactive power, can be specified. These target values ​​can be derived from the behavior of the DC link voltage and the load. Using the target apparent power, a desired system characteristic, such as the behavior of a constant resistance at the terminal or a constant power at the terminal, can be easily specified. It goes without saying that other values, such as a nominal voltage, a nominal current and the like, can be taken into account to achieve a specific system characteristic. A corresponding calculation of a setpoint can be performed (exclusively) in the time domain. By always assuming the system state to be steady-state, the setpoints can be converted into instantaneous target values. A predefined system characteristic can thus be implemented instantaneously in a simple manner. An optional maximum value correction can be used to ensure compliance with design-specific limitations. For this correction, the instantaneous current, i.e., the current flowing through the system, can be provided, particularly by measurement. This current can then be evaluated for the purpose of the maximum value correction. Preferably, the first setpoint indicator signal can be a first current setpoint indicator signal. Furthermore, the second setpoint indicator signal can be a second current setpoint indicator signal. In particular, the currents to be realized can be determined as the second current setpoint indicator signal. It is understood that the setpoint indicator signals can be time-dependent signals. According to a preferred embodiment, a prior first setpoint indicator signal or a prior second setpoint indicator signal is provided. The prior first setpoint indicator signal and / or the prior second setpoint indicator signal is rotated by an angle, the angle being determined as a function of the ascertained current time parameter. The rotated setpoint indicator signal and the first setpoint indicator signal are summed, and / or the rotated setpoint indicator signal and the second setpoint indicator signal are summed. In particular, the setpoint indicator signals can be current setpoint indicator signals. The rotation by an angle can correspond to the rotation of the pointer setting signal. Preferably, the signal from the two setpoint indicator signals can be summed with the pre-rotated setpoint indicator signal, which has been previously fed back. For example, if the second setpoint indicator signal is fed back, then the second setpoint indicator signal can be summed with the pre-rotated setpoint indicator signal. Preferably, the second current setpoint indicator signal is used. The reason for this is that the second current setpoint indicator signal has already undergone a maximum value correction. In particular, a negative pre-rotated current setpoint indicator signal can be summed with the new current setpoint indicator signal. A correction of the setpoint indicator signal can be carried out in a simple manner. As previously described, a setpoint indicator signal can be a current setpoint indicator signal. In particular, according to a further embodiment, the summed setpoint indicator signal can be a current setpoint indicator signal. From the current setpoint indicator signal, a current voltage pointer signal can be determined as the current pointer control signal using differential feedforward control. If summation is not performed, it is understood that the first or second current setpoint indicator signal can also be used instead of the summed current setpoint indicator signal. Differential feedforward control allows for the simple generation of a suitable instantaneous control signal, in particular a voltage pointer control signal. Furthermore, according to another embodiment, the current voltage pointer setting signal and the rotated voltage pointer setting signal can be summed. This allows for simple error correction. The resulting voltage pointer control signal can be supplied to a pulse generator. This can then control a power converter accordingly. Preferably, the (entire) control procedure is carried out in the time domain. In particular, an algorithmic procedure is provided according to the invention. Another aspect of the present invention is a self-commutated power converter, controlled according to the previously described method, with a control circuit for determining at least one current electrical parameter of a system. The control circuit includes a processing unit for determining at least one current time parameter from the current electrical parameter, wherein at least one previous pointer position signal is provided. The previous pointer position signal was determined in a previous position interval. The control circuit includes a rotary operator for rotating the previous pointer position signal by an angle, the angle being determinable as a function of the determined current time parameter. The control circuit is designed to control the power converter as a function of the rotated pointer position signal. Another aspect of the invention is a system with a previously described self-commutated converter, wherein the self-commutated converter is connected to a load and a grid. Another aspect is an energy generation plant, in particular a renewable energy generation plant, comprising the self-commutated power converter described above. The aforementioned methods can also be implemented as a computer program or as a computer program stored on a storage medium. In this case, a microprocessor can be appropriately programmed on the acquisition device, actuator, and / or server side to execute the respective process steps using a computer program. The features of the methods and devices can be freely combined with one another. In particular, features of the description and / or the dependent claims, even by completely or partially circumventing features of the independent claims, can be independently inventive, either on their own or freely combined. There are now numerous possibilities for designing and further developing the inventive method for controlling a self-commutated converter, the inventive self-commutated converter, and the inventive system. Reference is made, on the one hand, to the claims subordinate to the independent claims, and on the other hand, to the description of exemplary embodiments in conjunction with the drawings. The drawings show: Fig. 1 a flowchart of a first exemplary embodiment of the inventive method, Fig. 2 a simplified block diagram of an exemplary embodiment of a system according to the invention, Fig. 3 a further simplified block diagram of a further exemplary embodiment of a system according to the invention, Fig. 4 a simplified equivalent circuit diagram of an exemplary embodiment of a system according to the present invention.Fig. 5 shows a current-voltage curve of a first embodiment with two pulse inverters and a steady-state operating point, Fig. 6 shows a further current-voltage curve of a further embodiment with two pulse inverters and a change of the operating point, and Fig. 7 shows a further current-voltage curve of a further embodiment with two pulse inverters and an asymmetrical operating point. It was initially determined that sufficient grid stability requires all power generation plants to exhibit specific characteristics at the connection point. If energy is to be drawn from the grid, the behavior of a constant resistance must be emulated. If energy is to be fed into the grid, a constant power output must be maintained at the connection point. Since this is an AC voltage system, the required system characteristics can each be extended to their complex equivalent. Therefore, at the connection point, either the behavior of a constant complex resistance or the behavior of a constant complex apparent power can be implemented. The aforementioned requirements are formulated mathematically below. Since a component representation of complex numbers algorithmically represents the implementation with the lowest computational effort, this is also specified. The following equation can be used to describe the behavior of a constant complex resistance. The following applies to the desired apparent power: Here, the quantities labeled "nom" are the nominal quantities. From equations (a) and (b), the constant complex resistance can be expressed by the equation. The desired target current is then given by For constant power at the connection port, the following equation can be used. From this, the complex conjugate target current can be derived. Therefore, the following applies to the desired target current: Fig. 1 shows a flowchart of an embodiment of a method for controlling self-commutated power converters according to the present invention. In a first step 101, at least one current electrical parameter of a line can be provided, such as an actual voltage and / or an actual current. In the next step 102, a current time parameter of the line, for example the sampling time and / or the mains frequency, can be determined from at least one electrical parameter. Furthermore, a prior electrical pointer positioning signal can be provided in step 103. The previous electrical pointer position signal can be rotated by an angle determined based on the calculated time parameter (step 104). Specifically, the previous pointer position signal, like a space vector from the last control interval, can be rotated by the angle by which an (ideal) three-phase voltage system with the calculated mains frequency would have changed within the sampling time. Thus, the current phase angle is determined, and the previous pointer position signal is rotated accordingly, assuming that the previous pointer position signal was a correct pointer position signal. In step 105, the power converter can be controlled or regulated depending on the rotated electrical pointer position signal. Before controlling or regulating the power converter, the rotated pointer position signal can be corrected, as explained in more detail below. The procedure can then continue with step 101. It is understood that the individual steps can be carried out (almost) in parallel or sequentially. Fig. 2 shows a simplified block diagram of an embodiment of a system 200 according to the invention. The system 200 comprises at least one self-commutated converter 202, which can be connected to a load 204 and a network 206. The converter 202 further comprises a control circuit 208, which is provided for controlling the converter 202. In addition to suitable processor and memory components, the control circuit 208 can also include sensors for detecting electrical parameters of the system. A possible embodiment of the system according to the invention is explained in detail below. Fig. 3 shows a simplified block diagram of an embodiment of a system 300 according to the invention. The depicted system 300 initially comprises a section 302 or control section 302. The section 302 includes, in particular, the load and / or the network. For example, the section 302 can include network filters, converter inductors, transformers, and DC link capacitors. At least one electrical parameter of the system 302 is provided to a first processing unit 304 for analysis. The current parameter can, in particular, be an (output) voltage. This can be measured, for example, by a sensor. In particular, the phase voltage can be measured. In the present embodiment, this is a three-phase example, so that the current voltage is represented by a time-dependent, i.e., instantaneous, space vector. From the time-dependent electrical parameter, in particular at least one time parameter of the system 302 can be determined. This time parameter can be, in particular, the current mains frequency fmains or the current sampling time Tsampling. Preferably, both time parameters fmains and Tsampling are determined. The two time parameters fmains and Tsampling can be provided to a rotary operator device 306. In addition to the current mains frequency fmains and the current sampling time Tsampling, the pointer signals are also provided to the rotary operator 306 in the present embodiment. The pointer signal is the previous pointer control signal, in particular a voltage pointer control signal. The term "previous" here means that the pointer control signal was used in the last, i.e., the (immediately) preceding control interval for controlling the power converter 316, in particular the pulse inverter 316. In other words, feedback of the voltage pointer control signal is performed here. Similarly, the previous second setpoint indicator signal can be fed back to a second current setpoint indicator signal. The rotary operator unit 306 is configured to rotate at least the previous pointer control signal by an angle that can be derived from at least one time parameter. In particular, the instantaneous space vector of the converter control voltage from the last control interval can be rotated (exactly) by the angle by which an (ideal) three-phase voltage system with the determined frequency fnetwork would have changed within the sampling time Tsampling. Similarly, the instantaneous space vector of the phase current setpoint can be advanced. The rotated pointer position signal, in particular the pre-rotated voltage pointer position signal, can be the input of a further processing unit 308. The further processing unit 308 can be configured, in particular, to ensure the required system characteristics. As described above, the behavior of either a constant complex resistance or a constant complex apparent power is preferably implemented at the connection port. From the input variables, in particular a target active power Ptarget, a target reactive power Qtarget, and the inverted previous voltage vector signal, a corresponding first target vector signal, in particular a corresponding first current target vector signal, can be determined. The target values ​​of the active and reactive power Ptarget and Qtarget are not usually explicitly specified as target values, since these result from the behavior of the DC link voltage and the load. Furthermore, nominal values ​​can be provided. In order to transfer the power setpoints Psetpoint and Qsetpoint to an instantaneous setpoint, the system state can always be assumed to be steady-state. In particular, the fact that a current space vector can be instantaneously decomposed into a parallel and an orthogonal component for a given voltage space vector can be exploited, so that the steady-state setpoints can be converted into an instantaneous relationship. The first current setpoint indicator signal can be provided to a further processing unit 310 for maximum value correction. This optional maximum value correction can serve to ensure compliance with design-specific limitations. For maximum value correction, the instantaneous current can be provided, and in particular measured. This current can then be evaluated for maximum value correction. As a result, in the present embodiment, a second setpoint indicator signal is available as the output signal, specifically a second current setpoint indicator signal which corresponds to the current setpoint to be realized. The new current setpoint indicator signal can then be summed with the reversed previous current setpoint indicator signal, in particular with the negative value of . To advantageously achieve frequency-neutral behavior of the current implementation of the three-phase voltage system, an additional processing unit 312 can be used for differential feedforward control. This ensures that slow frequency statics integrated into the distribution and interconnected networks are not affected. Alternatively, a differential controller can be provided to achieve greater robustness. The output voltage space vector generated by the differential feedforward control 312 can be supplemented with the pre-rotated voltage space vector. The resulting voltage vector control signal can be transmitted to a pulse generator 314, in particular a pulse width modulator (PWM). The non-ideal switching characteristics of the power electronic elements, such as current-dependent turn-on and turn-off times, current-dependent voltage drops and the like, can be compensated for by an optional power converter correction. The resulting switching times and words are transferred to an actuator 316. In the specific case of a three-phase power converter 316, this can be a pulse inverter 316. This executes the switching operations, which in turn act on the control system 302. According to the invention, the control of the power converter 316 is carried out (exclusively) with instantaneous, i.e., time-dependent, quantities. Instantaneous control can be provided in a simple manner. Fig. 4 shows a simplified equivalent circuit diagram of an embodiment of a system 400 according to the present invention. The system 400 shown in Fig. 4 can be used in particular to investigate an embodiment of the present method. System 400 comprises several pulse inverters 402 connected on the AC side. A resistive-inductive impedance Z is connected to each pulse inverter 402 on the AC side, through which the mains voltage measurement of the control concept of the present embodiment according to the present invention can be performed. Without this impedance, only the voltage directly at the pulse inverter 402 could be measured, and the information about the mains voltage would be missing for the control system. Following the measuring impedance Zm is a line impedance Zk, also assumed to be ohmic-inductive. The individual line impedances Zks are connected on the network side to an inductive star point. A constant voltage source 404 is arranged on the DC side of the pulse inverter 402. This is because the realization of an ideal three-phase voltage system is intended, and therefore changes in the intermediate circuit voltages are irrelevant. In reality, a deviation from the target value of the intermediate circuit voltage would result in a change in the active power target value. The simulation is performed in interval-based mode. In other words, the individual switching of the semiconductor valves is neglected. Instead, the control output, and thus the AC voltage of each string of the converters 402, is kept constant over a control interval. This simplification has no effect on the fundamental frequency behavior of the system 400. The sampling times of the control systems are set to a constant Tsample = 100 µs for all 402 power converters in all configurations. The resistive-inductive impedances in this example are all assumed to be L = 10 mH and R = 100 µΩ. The same parameters were chosen for the measurement and line impedances Zs and Zk. This is an extremely demanding connection for the control structure according to the invention used in the present embodiment, since the relative short-circuit power is very low due to the large voltage drop. This unrealistic configuration was deliberately chosen to rule out random or borderline correct behavior of the control system. Under such extreme conditions, the focus is not on parameterization, but on principles and structure. The nominal voltage is UN = 1000 V, the current limit is Imax = 400 A. In order to clearly demonstrate the basic behavior, only two pulse inverters 402 are activated in the following embodiments. The setpoint values ​​for the apparent power are assumed to be Pset,1 = 300 kW and Qset,1 = -120 kVA, and Pset,2 = -Pset,1 and Qset,2 = -Qset,1, respectively. In this first embodiment, an operating point was therefore selected in which the setpoint values ​​of the fed-in and drawn-off apparent power correspond to each other. In other words, a suitable operating point was chosen. Figure 5 shows the currents iAC1,PWR1 and iAC1,PWR1 and the voltage uAC1,PWR1 over time. As can be seen in Figure 5, the operating point is also correctly set by the control approach according to the invention. The apparent power values ​​are Pist,1 = 301.14 kW and Qist,1 = -117.4 kVA, and Pist,2 = -303.1 kW and Qist,1 = 113 kVA, respectively. The difference between the actual and target values ​​is within the tolerable range. Furthermore, the difference between the apparent power of the first and second pulse inverters is necessary because, although the gate powers of the systems match, additional reactive power must be supplied to compensate for the line impedance. In the next embodiment, the operating point changes at t = 0.5 s to Psoll,1 = -Psoll,2 = -420 kW and Qsoll,1 = -Qsoll,2 = -192 kVA. The corresponding current profiles iAC1,PWR1 and iAC1,PWR1 and the voltage uAC1,PWR1 are shown in Fig. 6. As can be seen in Fig. 6, the change in the operating point occurs within two periods without significant overshoot, despite the absence of a controller. The setpoints are achieved with a quality level comparable to that of the first embodiment. It should be noted that good results were also achieved in tests with three or more power converters 402, both at a steady operating point and when changing the operating point. To demonstrate the potential of the present control approach, five pulse inverters 402 can be activated at the inductive neutral point. The setpoints are selected as Psoll,1 = 300 kW, Psoll,2 = -100 kW, Psoll,3 = -80 kW, Psoll,4 = 40 kW, and Psoll,5 = -160 kW, or Qsoll,1 = -120 kVA, Qsoll,2 = 50 kVA, Qsoll,3 = 30 kVA, Qsoll,4 = 40 kVA, and Qsoll,5 = 0 kVA. In this scenario as well, a good setpoint sequence and an operating point change within two periods are achieved. As in the previous scenario, all setpoints were multiplied by Pneu = -1.4 · Palt and Qneu = 1.6 · Qalt when changing the operating point. The reaction of system 400 to unsuitable setpoint values ​​for apparent power is subsequently examined using a further embodiment with two pulse inverters 402. For this purpose, the setpoint values ​​Psoll,1 = 300 kW, Psoll,2 = -150 kW, Qsoll,1 = 100 kVA and Qsoll,2 = 0 kVA are set. The resulting current profiles iAC1,PWR1 and iAC1,PWR1 and the voltage uAC1,PWR1 are shown in Fig. 7. The operating points Pist,1 = 223.57 kW, Pist,2 = -222 kW, Qist,1 = 51.25 kVA and Qist,2 = -52.89 kVA are obtained. The present method according to the invention for controlling power converters corrects the operating points of the pulse inverters with the two characteristics constant power and constant resistance in such a way that both pulse inverters have the same deviation from the setpoint and thus all participants make the same contribution to stabilizing the system, i.e. the grid. Tests have shown that even with more than two pulse inverters, the system responds with the nearest possible operating point for each inverter. For example, the System 400 also responds with five pulse inverters 402, with the nearest possible operating point for each inverter 402. In this example, with setpoints Psoll,1 = 300 kW, Psoll,2 = -100 kW, Psoll,3 = -80 kW, Psoll,4 = 100 kW, and Psoll,5 = -80 kW, there is a difference of ΔP = 140 kW between the active power drawn and the power fed into the grid. For reactive power, with Qsoll,1 = -150 kVA, Qsoll,2 = 120 kVA, Qsoll,3 = 30 kVA, Qsoll,4 = 90 kVA, and Qsoll,5 = 0 kVA, there is a difference of AQ = 90 kVA. In a simulation, the operating points Pist,1 = 275.6 kW, Pist,2 = -130.45 kW, Pist,3 = -107.9 kW, Pist,4 = 70.8 kW, and Pist,5 = -107.2 kW are obtained. With a difference of [value missing], this is the expected curve, as the active power of each pulse inverter 402 is approximately 28 kW lower than specified. This deviation is due to the cable and measurement impedances. The reactive power values ​​are Qist,1 = -163.6 kVA, Qist,2 = 101.2 kVA, Qist,3 = 10.96 kVA, Qist,4 = 73.1 kVA, and Qist,5 = -18.9 kVA. With a difference of [value missing] per pulse inverter 402, a reasonable operating point for the system 400 is also obtained here. The present embodiments with two pulse inverters demonstrate that steady-state and dynamic operation can be achieved for suitable operating points with a balanced apparent power balance. The time-domain investigations show, in particular, that even unbalanced setpoint scenarios are manageable. A suitable operating point between the autonomous, grid-connected power electronic systems can be realized without communication.

Claims

Method for controlling a self-commutated converter (202, 316, 402), wherein at least one current electrical parameter of a system (302) is provided, wherein: - at least one current time parameter of the system (302) is determined from the current electrical parameter, - at least one previous pointer control signal is provided, wherein the previous pointer control signal was determined in a previous control interval, - the previous pointer control signal is rotated by an angle, wherein the angle is determined as a function of the determined current time parameter, and - the converter (202, 316, 402) is controlled as a function of the rotated pointer control signal, - wherein a current pointer control signal is determined from the rotated pointer control signal, wherein the current pointer control signal is determined at least as a function of the rotated pointer control signal and at least one setpoint parameter to realize a specific system characteristic.- wherein at least one setpoint parameter is a setpoint apparent power, - wherein a first setpoint indicator signal is determined from the rotated indicator position signal and the setpoint apparent power, - wherein a second setpoint indicator signal is determined from the current of the line (302) and at least the first setpoint indicator signal by a maximum value correction, characterized in that - a previous first setpoint indicator signal or a previous second setpoint indicator signal is provided, - the previous first setpoint indicator signal and / or the previous second setpoint indicator signal is rotated by an angle, the angle being determined as a function of the determined current time parameter, and - the rotated setpoint indicator signal and the first setpoint indicator signal are summed, and / or - the rotated setpoint indicator signal and the second setpoint indicator signal are summed. Method according to claim 1, characterized in that the at least one current time parameter is a current network frequency of the line (302) and / or a current sampling time of the line (302), wherein the determined angle essentially corresponds to the angle by which the previous pointer setting signal has rotated from the previous setting time to the current setting time with the current network frequency and / or the current sampling time in an ideal rotating network. Method according to claim 1 or 2, characterized in that - the current voltage of the path (302) is determined as the current electrical parameter, and / or - the current current of the path (302) is determined as the current electrical parameter. Method according to one of the preceding claims, characterized in that - the summed setpoint indicator signal is a current setpoint indicator signal, and - a current voltage pointer signal is determined from the current setpoint indicator signal as the current pointer setting signal by means of a differential feedforward control. Method according to claim 4, characterized in that the current voltage pointer setting signal and the rotated voltage pointer setting signal are summed. A self-commutated power converter (200, 316, 402) controlled according to the method of one of the preceding claims, comprising a control circuit (208) for determining at least one current electrical parameter of a system (302), wherein: - the control circuit (208) has a processing device (304) for determining at least one current time parameter from the current electrical parameter, - wherein at least one previous pointer position signal is provided, the previous pointer position signal being determined in a previous position interval, - the control circuit (208) has a rotary operator device (306) for rotating the previous pointer position signal by an angle, the angle being determinable depending on the determined current time parameter, and - the control circuit (208) is provided for controlling the power converter (202, 316, 402) depending on the rotated pointer position signal. System (200, 300, 400) with a self-commutated converter (202, 316, 402) according to claim 6, wherein the self-commutated converter (202, 316, 402) is connected to a load (204) and a network (206).

Images