Method and control unit for reducing harmonic power flows, and sub-network having control unit

Abstract

The invention discloses a method for reducing harmonic power flows via a connection point (AP) at which a sub-network (10) is connected to a higher-level AC supply network (12) and via which a network current (I_Netz) flows between the sub-network (10) and the higher-level AC supply network (12). The sub-network (10) has at least one load (14) which draws, from the sub-network (10), an electrical load current (I_Last) which comprises an active power current (I_d) at a network frequency (f0) and a harmonic distortion current at one or more integer multiples of the network frequency (f0). The sub-network (10) also has a power converter (16, 16.N) which, by means of a bridge circuit, exchanges electrical power between a capacitor (18, 18.N) connected on its DC side and the sub-network (10) connected on its AC side. Firstly, in the method, a network voltage (Uac) of the sub-network (10) is detected. Subsequently, a compensation voltage and / or a compensation current (Komp) is determined using the network voltage (Uac), wherein the compensation voltage and / or the compensation current (Komp) is intended to be produced by means of the power converter (16, 16.N) and is suitable for reducing the harmonic distortion current at at least one multiple of the network frequency (f0) in the network current (I_Netz). The method also discloses producing the compensation voltage and / or the compensation current (Komp) by means of the power converter (16, 16.N) by suitable clocking of the bridge circuit between the DC-side capacitor (18, 18.N) and the sub-network (10).

Description

[0001]22-438-P-WO - 1 - submitted version METHOD AND CONTROL UNIT FOR REDUCING HARMONIC POWER FLOWS AND SUB-GRID WITH CONTROL UNIT TECHNICAL FIELD The application relates to a method and a control unit for reducing harmonic power flows as well as an AC sub-grid (AC sub-grid) connected to a higher-level AC supply network (AC supply network) with such a control unit. PRIOR ART AC sub-grids can be used in the industrial sector, e.g., to supply industrial plants and / or industrial areas. An AC sub-grid can consist of one or more AC cable strands that are connected together at a connection point and to which electrical loads can be connected. The loads connected to the sub-grid are supplied with electrical power with a current component at a nominal frequency of the sub-grid, e.g.at 50 Hz mains frequency and can also draw harmonic current components from the sub-grid, i.e. currents with multiples of the mains frequency. These are generally caused by consumers and equipment that have a non-sinusoidal current or a periodically switched on and off current flow, such as rectifiers, frequency converters or UPS systems and / or similar, particularly clocked loads. Multiples of up to 50 times the mains frequency can be observed. The result of the current distortion caused by these harmonics is a distortion of the nominally sinusoidal mains voltage with comparable frequency components to the harmonic current components. The voltages and / or currents in sub-grids for industrial plants and other larger objects with a high demand for electrical power, e.g.Production facilities with machine parks or shopping centers and the like contain distortions that exceed certain limits with regard to THD (total harmonic distortion) and / or harmonics in voltage and / or current. Such sub-grids and / or certain consumers supplied via individual strands of the sub-grid may therefore not be connected to a higher-level supply network without further ado if the aforementioned limits are defined for such a connection point. Passive filters are known which use passive elements such as capacitances and inductances. Such filters are designed, for example, as so-called absorption circuits and are usually designed or dimensioned for a specific load with given properties.22-438-P-WO - 2 - submitted version EP2436092 describes the compensation of harmonics occurring in current flows in a high-voltage grid using a passive filter and a controllable voltage source. A STATCOM is a power converter designed to exchange inductive or capacitive reactive power with an AC grid. A DC capacitor acts as a DC voltage source, which, via a power converter, forms an AC voltage source for connection to the AC grid via a transformer. EP2478610 describes a photovoltaic system connected to an AC supply grid via a transformer and designed to generate active power and exchange it with the AC supply grid using current control, as well as to compensate for reactive power and harmonics in the AC supply grid, so that the photovoltaic system operates as a type of STATCOM with current control.DE 10244056 B3 discloses a method for generating a set of control signals for a converter of an active filter for compensating harmonics. In this method, the current containing the harmonic to be compensated is measured, and the fundamental component is eliminated from the measurement signals. A transformed control function of a PI controller is applied to the measured values ​​filtered in this way. The output variables of the PI controller are then summed and transmitted as a setpoint current value to the control device of an IGBT converter. OBJECTIVE The application is based on the objective of providing a method and a control unit with which the quality of a sub-grid can be improved with regard to the electrical parameters within the sub-grid and / or at a connection point of the sub-grid to a higher-level supply grid.SOLUTION The problem is solved by a method having the features of claim 1 and a control unit having the features of claim 11. Embodiments are specified in the dependent patent claims. DESCRIPTION OF THE INVENTION A sub-grid, in particular an AC sub-grid, is connected to a higher-level AC supply grid at a connection point. A grid current flows via the connection point between the sub-grid and the higher-level AC supply grid, in particular from the higher-level AC supply grid into the sub-grid. The AC supply grid is higher than the sub-grid because, for example, it transports electrical energy over greater distances than the sub-grid, the grid voltage in the higher-level AC supply grid is higher than in the sub-grid and / or the AC supply grid is provided by a grid operator for a larger area to supply energy with defined grid parameters.The sub-grid has at least one electrical load that draws an electrical load current from the AC sub-grid, comprising an active power current at a grid frequency and a harmonic distortion current at one or more integer multiples of the grid frequency. The at least one electrical load can be, for example, an AC load, such as a motor, or a DC load that is connected to the AC sub-grid via a rectifier, for example. Depending on the power requirement of the load or the mode of operation of the converter, the above-mentioned components of the load current can be generated, for example, by a phase-angle control, via which the electrical power of an AC load can be adjusted.The sub-grid further comprises a power converter which, by means of a bridge circuit, exchanges electrical power between a direct current unit, in particular a capacitor, connected on its DC side and the sub-grid connected on its AC side. For this purpose, the bridge circuit can in particular comprise semiconductor power switches which are controlled in a clocked manner. The power converter can be operated bidirectionally and can therefore act as an inverter with electrical power flow from the DC side to the AC side and as a rectifier with electrical power flow from the AC side to the DC side. A method for reducing harmonic power flows via the connection point comprises the steps: - Detecting a grid voltage of the sub-grid.- Determining a compensation voltage and / or a compensation current using the grid voltage, wherein the compensation voltage and / or the compensation current is provided for generation by the power converter and is suitable for reducing the harmonic distortion current at at least a multiple of the grid frequency in the grid current. - Generating the compensation voltage and / or the compensation current by the power converter by suitable timing of the bridge circuit between the DC-side capacitance and the sub-grid. 22-438-P-WO - 4 - submitted version The sub-grid can be galvanically isolated from the higher-level AC supply grid. The galvanic isolation can be achieved in particular by a transformer at the connection point. In addition to the galvanic isolation, the transformer can effect a transformation of the voltage between the AC supply grid and the sub-grid.Galvanic isolation at the connection point is an important safety feature for connection to AC supply networks. The sub-network can be designed in such a way that it does not require internal galvanic isolation. In particular, the power converter can be connected to the sub-network without galvanic isolation. The described method can improve the grid voltage quality in the sub-network. This can be particularly advantageous for industrial applications that may have loads that negatively influence the grid voltage. The method can improve the grid current quality and thus the grid voltage via the grid impedance. Both quality improvements can have a targeted effect on improving the power exchanged with the higher-level AC supply network via the connection point and reducing the negative effects of the power exchange via the connection point on the higher-level AC supply network.On its DC side, the power converter can alternatively or additionally be connected to an electrical DC source, such as a PV system (PV: photovoltaics). The power converter can draw electrical power from the DC source in inverter mode and feed it into the sub-grid. This electrical power drawn from the DC source can be used, for example, to cover the power demand in the sub-grid. This is advantageous in the industrial sector, for example, because it can maximize the self-consumption of locally generated power. In addition, a grid-friendly compensation function can be implemented by providing distortion reactive power to smooth the current at the connection point. This grid-friendly compensation function can be prescribed by regulation and / or performed at the request of the AC supply grid and / or the grid operator.With the aid of the claimed method, it is possible to meet this requirement and at the same time reduce the effort required for any AC line filters, which enables cost reductions. Within a sub-network, a compensation voltage and / or a compensation current is provided locally, which provides the required harmonic distortion current to the at least one load, so that the harmonic distortion current to be provided by the AC supply network and thus the harmonic power flows via the connection point are reduced. 22-438-P-WO - 5 - filed version. In embodiments of the method, different harmonics of the harmonic distortion current at different multiples of the line frequency can be considered independently of one another when determining the compensation voltage and / or the compensation current.In one embodiment of the method, a harmonic controller uses the mains voltage to determine a respective contribution of the respective harmonic to the mains voltage. Each respective harmonic contribution is an oscillation of the mains voltage with a frequency that is a respective multiple of the mains frequency, wherein this oscillation of the mains voltage is electrically coupled to a corresponding oscillation of the distortion current. The harmonic controller uses the respective harmonic contribution as a controlled variable to determine a respective harmonic compensation contribution to the compensation voltage. In particular, odd multiples of the mains frequency can be used here. It is further proposed that a DC voltage is detected that is present across the capacitance connected on the DC side of the power converter.A DC control system determines a DC control contribution to the compensation voltage and / or the compensation current using the DC voltage in such a way that the DC voltage is regulated to a predefined level through a suitable active power exchange via the bridge circuit. This control system compensates for any active power exchange due to the other contributions to the compensation voltage or current, in particular, active power due to the harmonic compensation amounts, without interfering with the control of the other contributions, so that the capacity of the power converter is permanently available as a source for the compensation voltage or current.In embodiments, the DC control comprises an f(P)-PI controller (PI controller = proportional-integral controller), which uses as an input value a difference between an active power current setpoint and an active power current actual value of the active power exchanged via the bridge circuit. The active power current setpoint is determined as a function of a difference between an actual value of the DC voltage and the specified level of the DC voltage. The active power exchange is controlled via the bridge circuit by changing the AC-side frequency of the power converter as a function of the output value of the f(P)-PI controller. The AC-side frequency of the power converter corresponds to the output frequency of the electrical quantities present on the AC side of the power converter.Specifically, for example, the output frequency of the power converter can be changed by an amount of change composed of a first amount generated by the P component of the f(P) PI controller (22-438-P-WO - 6 - filed version) and proportional to the difference between the active power current setpoint and the active power current value, and a second amount generated by the I component of the f(P) controller and proportional to the time integral of the difference between the active power current setpoint and the actual value. The embodiment with the f(P) PI controller enables grid-forming control of the power converter by adjusting a phase angle difference between the output voltage of the power converter and the grid voltage by changing the AC-side frequency of the power converter.This allows, on the one hand, the active power current setpoint to be precisely adjusted, and, on the other hand, the control system reacts to a grid event, such as a phase shift in the grid voltage or a change in the grid frequency, at least temporarily, with a corresponding active power change. In an alternative embodiment, the DC control system comprises a P-controller (proportional controller) that uses a difference between an actual value of the DC voltage and the specified level of the DC voltage as an input value. The active power exchange via the bridge circuit is controlled by specifying a current setpoint depending on the output value of the P-controller. Specifically, for example, a current setpoint can be generated that is proportional to the difference between the actual value and the setpoint of the DC voltage.The design with a P-controller enables efficient DC voltage control within the characteristic linearized voltage range of the DC unit. Furthermore, DC control with a P-controller enables rapid adjustment of the actual value to the DC voltage setpoint. The load current can have a reactive power current at the grid frequency. To reduce the resulting reactive power exchange at the connection point of the subgrid to the higher-level AC supply grid, the method can include reactive power control. The reactive power control determines a reactive power control contribution to the compensation voltage. For this purpose, the reactive power control comprises a U(Q)-PI controller, which uses as input a difference between a reactive power current setpoint and an actual reactive power current value of the reactive power exchanged via the bridge circuit.The reactive power current setpoint is specified for the reactive power as a function of a difference between a setpoint and an actual value of the grid voltage such that the reactive power exchange at the connection point is reduced. The method can comprise the further steps: - detecting the load current and the grid current, - determining a grid current setpoint using the load current, - determining a setpoint output voltage of the power converter as a function of the difference between an actual grid current value and the grid current setpoint, - generating a control signal from the determined setpoint output voltage and clocking the bridge circuit as a function of the control signal in order to generate the compensation current. The control signal can in particular be a pulse width modulation signal which specifies the clocking of the semiconductor switches of the bridge circuit, e.g.by specifying suitable opening and closing times for the semiconductor switches. Determining the grid current setpoint can comprise bandpass filtering of the load current, wherein the center frequency of the bandpass filter depends on the grid frequency. The center frequency of the bandpass filter can essentially correspond to the grid frequency. In embodiments of the method, the load current, the output current of the power converter, and / or the grid voltage can be recorded and used to pre-control the output voltage of the power converter. The sub-grid, which is connected to the higher-level AC supply grid via the connection point, has at least one load that draws the load current from the sub-grid. The load current comprises an active power current at a grid frequency and a harmonic distortion current at one or more integer multiples of the grid frequency.The sub-grid also has a power converter which, by means of a bridge circuit, exchanges electrical power between a capacitor connected on its DC side and the sub-grid connected on its AC side.A control unit for reducing harmonic power flows in the grid current flowing via the connection point between the sub-grid and the higher-level AC supply grid is designed to: - receive a grid voltage of the sub-grid, - determine a compensation voltage and / or a compensation current using the grid voltage, wherein the compensation voltage and / or the compensation current is provided for generation by the power converter and is suitable for reducing the harmonic distortion current at at least a multiple of the grid frequency in the grid current, and - output a control signal to the power converter, wherein the compensation voltage and / or the compensation current can be generated by the power converter using the control signal by suitable clocking of the bridge circuit between the DC-side capacitance and the sub-grid.22-438-P-WO - 8 - filed version. In one embodiment, the control unit is configured to independently determine a respective harmonic compensation contribution to the compensation voltage for different harmonic contributions at different multiples of the grid frequency. The control unit can be configured to implement a respective harmonic controller, by means of which a respective harmonic contribution can be determined using the grid voltage, wherein the respective harmonic compensation contribution to the compensation voltage can be determined using the respective harmonic contribution as a controlled variable.In embodiments, the control unit can be configured to receive a DC voltage applied to the capacitor connected on the DC side of the power converter and to execute a DC control, by means of which a DC control contribution to the compensation voltage and / or to the compensation current can be determined using the DC voltage such that the DC voltage is regulated to a predeterminable level by suitable active power exchange via the bridge circuit. The load current can further comprise a reactive power current at the grid frequency, and the control unit can be configured to execute a reactive power control, by means of which a reactive power control contribution to the compensation voltage can be determined such that the reactive power exchange at the connection point is reduced. The sub-grid that is connected to the higher-level AC supply grid at the connection point can have such a control unit.The sub-grid has at least one line section to which at least one load is connected, which is designed to draw the load current from the sub-grid, which includes the active power current at the grid frequency and the harmonic distortion current at one or more integer multiples of the grid frequency. In embodiments, the load current can also include the reactive power current at the grid frequency. The sub-grid further has the power converter, which is designed to exchange electrical power between the direct current unit connected on its DC side, in particular a capacitor, and the sub-grid connected on its AC side, by means of the bridge circuit, in such a way that harmonic power flows via the connection point are reduced.The sub-grid may comprise multiple power converters, particularly on the same power line of the sub-grid, wherein the various power converters are each configured to generate a compensation voltage and / or a compensation current to reduce different harmonic contributions at different multiples of the grid frequency. 22-438-P-WO - 9 - submitted version. The power converters may be connected to a higher-level control unit. Alternatively or additionally, corresponding control units may be arranged decentrally on individual or all power converters in the sub-grid.The higher-level or decentralized control unit can output a respective control signal to a respective one of the multiple power converters, wherein the respective control signal is suitable for generating the compensation voltage and / or the compensation current by the respective power converter to reduce the harmonic distortion current at one or more multiples of the grid frequency. If the load current has a reactive power current at the grid frequency, the respective control signal can be suitable for generating the compensation voltage and / or the compensation current by the respective power converter to reduce the reactive power exchange at the connection point. The sub-grid can be galvanically isolated from the higher-level AC supply grid, in particular by a transformer at the connection point.The sub-network itself can be designed without galvanic isolation, and in particular the at least one power converter can be connected to the sub-network without galvanic isolation. BRIEF DESCRIPTION OF THE FIGURES The technical teaching according to the invention is further explained and described below with reference to exemplary embodiments shown in the figures. Fig. 1 schematically shows a method for reducing harmonic power flows. Fig. 2 schematically shows an exemplary embodiment of a sub-network with a power converter. Fig. 3 schematically shows an exemplary embodiment of a control system with various components. Fig. 4 schematically shows a further exemplary embodiment of a sub-network with a power converter. Fig. 5 schematically shows a further exemplary embodiment of a sub-network with a power converter. Fig. 6 schematically shows a further exemplary embodiment of a control system with various components. Fig.7 shows schematic example current and voltage curves in the sub-grid. 22-438-P-WO - 10 - filed version Fig. 8 shows schematically a further embodiment of a sub-grid with a power converter and DC source. Fig. 9 shows schematically a further embodiment of a sub-grid with multiple power converters. The same reference numerals are used in the figures for the same or similar elements. The representations in the figures may not be to scale. DESCRIPTION OF THE FIGURES Fig. 1 shows schematically a method for reducing harmonic power flows via a connection point AP, at which a sub-grid 10 with a power converter 16, 16.N is connected to a higher-level AC supply network 12. The method has the steps: S1: Detecting a grid voltage U. ac of the sub-network 10. S2: Determination of a compensation voltage and / or a compensation current Komp using the grid voltage U ac. S3: Generation of the compensation voltage and / or the compensation current Komp by the power converter 16, 16.N. The method is carried out repeatedly, in particular, so that distortions caused by harmonics in the electrical power exchanged via the connection point can be reduced by the continuous generation and adjustment of the compensation current and / or the compensation voltage Komp. The method is carried out, for example, by a control unit 20, 20.N designed as a computing unit with memory and processor. The method can be executed on the control unit 20, 20.N, e.g., as software. Appropriate measuring devices can be provided for recording measured values. The compensation voltage and / or the compensation current Komp are determined by means of the method, e.g., by the control unit 20, such that they can be generated by the power converter 16, 16.N. Fig.2 schematically shows an embodiment of the sub-network 10 with the power converter 16. In the example shown, the sub-network 10 is designed as an AC sub-network with one line harness. In the example shown, the sub-network 10 has three loads 14. More or fewer loads 14 are also conceivable. The loads 14 can in particular be designed as AC loads or as DC loads, which are connected to the sub-network 10 directly or via suitable converters. The sub-network 10 is connected to the AC supply network 12 via the connection point AP. A network current I_Network flows between the sub-network 10 and the higher-level AC supply network 12 via the connection point AP. The sub-network 10 can, for example, be designed so that it does not require any galvanic isolation within the sub-network 10. In particular, the power converter 16 can be connected to the sub-grid 10 without a transformer.The connection of the sub-grid 10 to the AC supply grid 12 via the connection point AP is designed with galvanic isolation, in particular by a transformer T. The loads 14 draw an electrical load current I_Load from the sub-grid 10. The load current I_Load comprises an active power current I_d at a grid frequency f0 and a harmonic distortion current at one or more integer multiples of the grid frequency f0. The load current I_Load can further comprise a reactive power current I_q at the grid frequency f0. The power converter 16 has an AC side and a DC side and can be operated bidirectionally, i.e. as an inverter and / or as a rectifier. The power converter 16 is configured to exchange electrical power between a capacitor 18 connected on its DC side and the sub-grid 10 connected on its AC side by means of a bridge circuit.Electrical power can be stored in the capacitor 18 and drawn from it again via the power converter 16. The power converter 16 can be connected to the sub-grid, for example, via a filter inductor 22. The control unit 20 measures the grid voltage U using a suitable measuring device. ac. and determines the compensation voltage and / or the compensation current Komp such that the compensation voltage and / or the compensation current Komp can be generated by suitable clocking of the bridge circuit of the power converter 16. For example, the suitable clocking of the bridge circuit can be generated via a control signal PWMS, e.g., by pulse width modulation. The compensation voltage and / or the compensation current Komp generated in this way by the power converter 16 is suitable for reducing the harmonic distortion current at at least a multiple of the grid frequency f0 in the grid current I_grid. To start the power converter 16, the DC voltage UDC can first be ramped up using a precharging circuit (not shown) supplied on the AC side.When the DC voltage UDC is sufficiently precharged, the actual operation of the power converter 16 to reduce the harmonic power flows is started, and the DC voltage UDC is adjusted by means of a DC control 26. Thereafter, during operation of the power converter 16 as described above, the grid voltage U can be harmonized by generating compensation voltage and / or compensation current Komp. accontributed. Fig.3 shows a schematic of an embodiment of a control system that can implement the described method. The control system can run, for example, in the control unit 20 and comprise various components. In the embodiment shown in Fig.3, the control system comprises three sub-aspects: harmonic control 24, DC control 26, and reactive power control 28, which can be active at the same time. The DC control 26 and reactive power control 28 described below operate in the so-called dq system. This results from a transformation from an abc system with time profiles of the individual phase voltages or phase currents into the dq system with symmetrical components and is regularly used for multi-phase alternating currents. The dq system regularly has the symmetrical components positive sequence system, negative sequence system and zero sequence system. The DC control 26 can ensure a constant DC voltage UDC.The DC voltage on the DC side, in particular the DC voltage of a capacitor 18 or several capacitors 18 of the DC voltage intermediate circuit, is stabilized by suitable power exchange via the power converter 16 with the sub-network 10 connected on the AC side, by generating a positive sequence current, i.e. an active current, I_d, which depends on the deviation of the actual value ^^^^. ^ ^ ^ ^^ ^ ^ ^ ^^ ^ ^ ^ ^ ^ ^ ^ ^^^ the DC voltage from a DC target voltage ^^^^ ^^ ^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^ ^^^^ ^^^^via an active current regulator ^^^^ ^^^^ ( Δ ^^^^ ^^^^ ^^^^ ) adds or removes energy from the DC link capacitances, e.g., capacitance 18. Using the difference between the actual value the DC voltage UDC and the specified level USolldc of the DC voltage UDC, the active power current setpoint Isoll d for the active power. The DC control 26 includes an f(P)-PI controller, which uses a difference ∆I as input value d between an active power current setpoint I d soll and an active power current actual value I d ist of the active power exchanged via the bridge circuit of the power converter 16. The active power exchange via the bridge circuit of the power converter 16 is controlled by changing the AC-side frequency f soll of the power converter 16 is controlled depending on the output value of the f(P)-PI controller. 22-438-P-WO - 13 - submitted version. The reactive power control 28 responds to the AC voltage amplitude �U ac �, i.e. the magnitude of the mains voltage U ac . With increased actual value�U a is c t � the voltage amplitude�U ac� is controlled by a reactive power current controller I q ( ΔU ac ) generated reactive power current I q Reactive power is absorbed. If the the voltage amplitude �U ac � is controlled by the reactive power current controller I q (ΔU ac ) generated reactive power current I q Reactive power is provided. If the connected loads 14 in sub-grid 10 are predominantly inductive, for example, motors, reactive power is provided by sub-grid 10 for their operation. This reactive power interacts particularly with inductances in sub-grid 10 and leads to an increased AC voltage amplitude �U ac � in sub-grid 10. If, however, the loads 14 are predominantly capacitive, reactive power is fed into sub-grid 10. This leads to a reduced AC voltage amplitude �U ac�. By providing corresponding counter-current reactive power, the power converter 16 thus relieves the sub-grid 10 from the reactive power exchange with the AC supply network 12 via the connection point AP. In the embodiment shown in Fig. 3, the reactive power is provided by the power converter 16 generating a negative sequence current I q which is then measured at the output of the power converter 16. The reactive power control 28 determines a reactive power control contribution to the compensation voltage Komp. The reactive power control 28 includes a U(Q)-PI controller, which receives as input a difference ∆ ^^^^ ^^^^ between a reactive power current setpoint ^^^^ ^^ ^^ ^ ^ ^^ ^^^^ ^^^^ ^^^^and a reactive power current actual value ^^^^ ^ ^ ^ ^ ^ ^^ ^ ^^^^ ^^^^the reactive power exchanged via the bridge circuit of the power converter 16. The reactive power current setpoint ^^^^ ^ ^ ^ ^ ^ ^ ^^ ^^^^ ^^^^ ^^^^for the reactive power is determined depending on a difference between a setpoint�^^^^ ^^^^ ^^^^ and an actual value� ^^^^ ^^^^ ^^^^ � the magnitude of the mains voltage� ^^^^ ^^^^ ^^^^ � specified. As an output value, the U(Q)-PI controller generates a voltage difference ∆ ^^^^ which is used together with a specified voltage U0 to generate a desired effective voltage U d soll Together with a target reactive power voltage U q soll from zero, the target voltage U d so q ll in the dq system. The output values ​​of DC control 26 and reactive power control 28 are transformed into the abc system and result in the grid-frequency target voltage U F So u l n l damental . There may be loads 14 that receive certain harmonics from sub-grid 10 via the load current I_Load. This causes a harmonic distortion current at one or more multiples of the grid frequency f0. The harmonic control 24 has a respective harmonic controller 24.3, 24.5, 24.n for harmonics at one or more multiples of the grid frequency f0. In the example shown in Figure 3, the harmonic control 24 has harmonic controllers for harmonics with three times the grid frequency f0, five times the grid frequency f0, and n times the grid frequency f0. Preferably, n is chosen as an odd natural number. A respective harmonic controller 24.3, 24.5, 24.n filters from the actual value the mains voltage U ac the respective harmonic contribution U3 is t t e Harm U5 is t t e Harm U nis t t e Harm This is used as a control variable and is adjusted by setpoints U3 which are set to zero. S t o e ll H arm = 0 for the harmonics. The harmonic control 24 generates the harmonic contributions U H So a l r l m onic to the compensation voltage. These are compared with the mains frequency target voltage U F S u ol n l d amental from the DC control 26 and the reactive power control 28 in the abc system combined, and the target voltage results ^^^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^ ^ ^^^^ in the abc system. Using the target voltage ^^^^ ^^ ^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^^ ^ ^^^^In the abc system, the compensation voltage Komp is generated by the power converter 16. For this purpose, the control signal PWMS is generated by generating the control signal PWM. The semiconductor switches of the power converter 16 are suitably controlled by means of the control signal PWMS, and the AC output voltage of the power converter 16 is suitably adjusted by a compensation voltage Komp. In the embodiment shown in Fig. 3, the compensation voltage Komp then subsequently generates a compensation current Komp. In the upper part of Fig. 4, time profiles of the mains current I_Netz and load current I_Last are graphically represented as examples. The compensation current Komp generated by the power converter 16 and symbolically represented at the output of the power converter 16 improves the quality of the mains current I_Netz exchanged via the connection point AP.In particular, harmonic power flows contained in the load current I_Load as harmonics are not provided via the connection point AP from the higher-level AC supply network 12, but rather by the power converter 16, so that the grid current I_Network essentially comprises a sinusoidal waveform with the grid frequency, and the compensation current Komp corresponds to the harmonic distortion current in the load current I_Load. Fig. 5 schematically shows another embodiment of the sub-network 10 with a line section and the power converter 16 connected to it. The computing unit 20 records the grid voltage U using suitable measuring devices. ac, the mains current I_Mains, a filter current I_Filter, the DC voltage UDC and the load current I_Load. The control unit 20 determines the compensation voltage and / or the compensation current Comp such that the compensation voltage and / or the compensation current Comp 22-438-P-WO - 15 - filed version can be generated by suitable clocking of the bridge circuit of the power converter 16. The control signal PWMS can be used, for example, to generate the suitable clocking of the bridge circuit, e.g., by pulse width modulation. In this exemplary embodiment, too, the harmonic distortion current of the load current I_Load can be supplied by the power converter 16 and does not have to be supplied, or not completely, from the AC supply network 12.The AC supply network 12 and the electrical power exchanged via the connection point AP can thus be relieved of the harmonic distortion current, so that the harmonic power flows via the connection point AP are reduced. For this purpose, the grid current I_grid is measured "before" and the load current I_load "after" the power converter 16. The power converter 16 acts like a filter that can "filter out" any harmonics that occur from the grid current I_grid without superimposed specifications. Fig. 6 shows a schematic illustration of an exemplary embodiment of a control system that implements the method according to the application and can run, for example, in the control unit 20. A DC control system 30 comprises a proportional controller (P controller) P, which has as input value a difference between an actual value. the DC voltage UDC and the specified level U d So c llof the DC voltage UDC. As an output value, the P-controller P outputs a DC current ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^, which is suitable for stabilizing and / or reducing the difference between the actual and setpoint values ​​of the DC voltage UDC, so that the DC voltage UDC is regulated to a preset level. Using the load current I_Last, a mains current setpoint I netzsoll in the time domain and transformed from the abc system to the dq system. This results in the grid current setpoint in the dq system I netzdqsoll . Determining the mains current setpoint I netzsoll may include a bandpass filtering of the load current I_Last in the time domain, where the center frequency of the bandpass depends on the mains frequency f0 and may correspond to the mains frequency f0. The resulting mains current setpoint I netzdqsollThis essentially corresponds to the desired sinusoidal curve of the mains current I_Mains at the mains frequency f, ie, the load current I_Load less any harmonic distortion currents. The measured mains current I_Mains is also transformed from the abc system to the dq system and results in the mains current I netzdq dq-System. Depending on the actual value Inetzdq of the mains current and the mains current setpoint Inetzdqsoll and the output value of the DC control 30, namely the DC current ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^, 22-438-P-WO - 16 - submitted version, first a current setpoint in the dq-System and then via a proportional-integral-derivative controller (PID controller) PID the setpoint output voltage U dqsoll of the power converter 16 in the dq system. Specifically, a difference between the mains current setpoint I netzdqsoll and mains current actual value I netzdqdetermined and added to the DC current ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ . This sum is then fed as the current setpoint in the dq system IWRdqsoll to the PID controller PID, which provides the setpoint output voltage U at its output. dqsoll outputs a compensation current Komp that is suitable for generating a compensation current Komp, which includes both any harmonic distortion currents and any active currents for regulating the DC voltage UDC. The control further comprises a feedforward control circuit that can be used to pre-control the output voltage U_dq of the power converter 16. The input value of the feedforward control circuit is the target output voltage U dqsoll of the power converter 16. The other input variables include the load current I_Last, which was transformed from the abc system into the dq system, the mains voltage U ac , which was transformed from the abc system into the dq system, the frequency of the mains voltage U acand the output current I_filter of the power converter 16. Applying these variables to the target output voltage Udqsoll allows for changes in the various control input variables to be taken into account, even before these changes have a concrete effect on the target output voltage along the controlled system. In particular, the feedforward control using the output current I_filter as well as the load current I_Last and the mains voltage U accan enable a faster response of the control to, for example, externally induced changes in the output current I_filter of the power converter 16. In addition, the feedforward compensation compensates for the influence of periodically fluctuating AC variables on the DC control, so that a P-controller is sufficient for the DC control, whereby the dynamic behavior of the DC voltage can be further improved by means of a high proportionality factor. The output value of the feedforward compensation is the output voltage U_dq of the power converter 16. The output voltage U_dq of the power converter 16, which is subsequently transformed from the dq system to the abc system, is used by the control signal generator PWM to generate the control signal PWMS for the bridge circuit of the power converter 16. The bridge circuit is clocked as a function of the control signal PWMS in order to generate the compensation current Komp.In the embodiment shown in Figures 5 and 6, the compensation current Komp is linked to a compensation voltage Komp, which results in the control according to Figure 6 such that the filter current I_filter fed in by the power converter 16 includes both any harmonic distortion currents and any active currents for regulating the DC voltage UDC. 22-438-P-WO - 17 - filed version Figure 7 shows exemplary schematic representations of the mains current I_Netz, load current I_Last, output current of the power converter I_filter, and the DC voltage UDC. It can be seen that the load current I_Last corresponds to the mains current I_Netz via the connection point AP when the power converter 16 is inactive. This corresponds to the exemplary representation in Figure7 the period t < t1, in which both the load current I_Lost and the grid current I_Grid include a harmonic distortion current in addition to a grid-frequency fundamental wave, while the filter current I_filter is zero. At time t1, the power converter 16 is activated and executes a method according to the application for reducing the harmonic power flows in the grid current I_Grid. In a short transient phase, the DC voltage UDC drops briefly and is raised back to a normal level by the DC control used. In the steady state in the period between t1 and t2, the harmonic distortion currents in the load current I_Lost are taken over by the power converter 16 and fed in as filter currents I_filter, so that the grid current I_Grid essentially only includes the sinusoidal grid-frequency fundamental wave.At time t2, the load current I_Load changes, and another short transient phase follows with a regulation of the briefly dipping DC voltage UDC. The transient phase can be shortened in particular by means of the feedforward control according to Fig. 6. The harmonic power flows at the connection point of the AP are subsequently reduced again by the harmonic distortion currents in the load current I_Load being taken over by the power converter 16 and fed in as filter currents I_filter. Fig. 8 shows an embodiment in which the power converter 16 has a connected DC source G. The previously described methods and controls can be supplemented here by the feeding in of active electrical power from the DC source G. For example, the power converter 16 can carry out an AC current-controlled feed-in from a PV system that acts as a DC source G, with a selective grid forming function.A power converter 16 designed as a PV inverter 16 can feed PV power into the sub-grid 10 in a voltage-controlled manner, e.g., grid-following / current-impressing / current-regulated, in the form of AC active power, and can also, within the scope of its capabilities, specifically influence certain harmonics, e.g., by means of voltage-impressing harmonic control as shown in Fig. 3 or current-impressing filtering as shown in Fig. 6. The active power of a PV generator can be fed into the AC supply grid via AC current-controlled operation at the grid frequency f0. In addition, the power converter 16 can feed a compensation voltage and / or a compensation current 22-438-P-WO - 18 - submitted version Komp into the sub-grid 10, the frequency of which compensates for one or more AC harmonics, e.g. B. the 3rd, 5th, 7th, 9th, 11th, 13th harmonics, etc., with the aim of minimizing these AC harmonics at the connection point AP to the AC supply network 12.This allows the power converter 16, e.g., PV inverter 16, to be specifically used as an active grid filter. Furthermore, a decision can be made as to whether the permissible apparent power of the power converter 16 should be used to feed the full power of the DC source G into the AC grid as active power, or whether parts of the permissible apparent power of the power converter 16 should be used to reduce the distortion current in the grid current I_grid. Such a distortion current can, for example, be generated by a load 14, which can, for example, be a large consumer among the consumers 14. As a result, the power converter 16 can contribute to complying with limit values ​​with regard to THD (Total Harmonic Distortion) and / or harmonics, particularly at the connection point AP, which would otherwise be exceeded by the operation of the load 14 in the form of the large consumer. Such behavior can, for example,be suitably remunerated by the large consumer and / or the grid operator of the AC supply grid. Fig. 9 shows an embodiment of the sub-grid 10 with several power converters 16, 16.N. The power converters 16, 16.N can have capacitors 18, 18.N connected to their respective DC sides, to which DC voltages UDC, UDC.N are applied. On the AC side, the power converters 16, 16.N can be connected to the sub-grid 10 via filter inductors 22. Each power converter 16, 16.N can have a control unit 20, 20.N. A higher-level control unit is also conceivable which, alternatively or additionally, controls both and possibly further power converters 16, 16.N, for example by specifying individual parameters of the controls in the power converter 16, 16.N, and / or executes the previously described methods and controls. The power converters 16, 16.N can be connected in parallel on the AC side of subnetwork 10. Different subfunctions, e.g.The sub-aspects concerning harmonic control 24, DC control 26, 30, and reactive power control 28 can be fulfilled by different power converters 16, 16.N. It is also conceivable for multiple power converters 16, 16.N to fulfill the same sub-function. In particular, a first number of power converters 16, 16.N can take over the reduction of harmonics in the grid current I_Grid, with each power converter 16, 16.N being able to reduce a specific harmonic; at the same time, a second number of power converters 16, 16.N can perform reactive power compensation. This reduces the load on the power converters 16, 16.N and can be protected, thereby increasing their service life. 2-438-P-WO - 19 - submitted version LIST OF REFERENCE SYMBOLS Subnetwork 12 AC supply network Load 16, 16.N Power converter 18, 18.N Capacitance 20, 20.N Control unit 22, 22.N Filter inductances 24 Harmonic control 24.3, 24.5, 24.n Harmonic controller 26 DC control Reactive power control 30 DC control Comp Compensation voltage / compensation current f0 Mains frequency. ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ Target frequency∆ ^^^^ Voltage difference ^^^^0Specified voltage ^^^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^^^^ ^^^^ Target voltage dq system^^^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^ ^^^^^ Target voltage abc-SystemAP Connection point G DC source 2-438-P-WO - 20 - submitted version P Proportional controller PI Proportional-integral controller PID Proportional-integral-derivative controller PWM Control signal PWMS, PWMS.N Control signal generation S1, S2, S3 Process steps T Transformer UDC, UDC.N DC voltage ^^^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^^ ^^^^ ^^^^specified level of DC voltage D C-Spannung (Istwert) ^^^^ ^^^^ (Δ ^^^^ ^^^^ ^^^^ ) Active current regulator ^^^^ ^^ ^^ ^ ^ ^^ ^^^^ ^^^^ ^^^^ Active power current setpoint^^^^ ^ ^ ^ ^ ^ ^^ ^^^^^ ^^^^ Active power actual current value∆ ^^^^ ^^^^ Difference between active power and current values ​​∆f frequency difference USoll Fundamental Target fundamental voltageUac Mains voltage 2-438-P-WO - 21 - submitted version I_q Reactive power current I q (ΔU ac ) Reactive power controller Iq target reactive power current Iq is actual reactive power current ∆Iq Difference reactive power current ∆u voltage difference Usoll d Target effective voltage Usoll q Target reactive power voltage Ud so qll Target voltage in the dq system UH So a l r l m onicHarmonic setpoint voltageI_Netz Mains current Ifilter Filter current I_Lost Load currentInetzdq Mains current dq-SystemInetzsoll Mains current setpointInetzdqsoll Mains current setpoint dq-SystemIWRdqsoll Current setpoint dq-SystemU_dq Output voltage power converter dq-System U dqsoll Target output voltage dq-System ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^DC current

Claims

22-438-P-WO - 22 - filed version PATENT CLAIMS 1. Method for reducing harmonic power flows via a connection point (AP) at which a sub-grid (10) is connected to a higher-level AC supply grid (12) and via which a grid current (I_grid) flows between the sub-grid (10) and the higher-level AC supply grid (12), wherein the sub-grid (10) has at least one load (14) which draws an electrical load current (I_load) from the sub-grid (10), which load current comprises an active power current (I_d) at a grid frequency (f0) and a harmonic distortion current at one or more integer multiples of the grid frequency (f0), wherein the sub-grid (10) further has a power converter (16, 16.N) which, by means of a bridge circuit, converts electrical power between a capacitor (18, 18.N) and the sub-grid (10) connected on its AC side, and wherein the method comprises the steps of: - detecting a grid voltage (U. ac ) of the sub-grid, - determining a compensation voltage and / or a compensation current (Komp) using the grid voltage (U ac), wherein the compensation voltage and / or the compensation current (Komp) is suitable for reducing the harmonic distortion current at at least one multiple of the grid frequency (f0) in the grid current (I_Netz), and - generating the compensation voltage and / or the compensation current (Komp) by the power converter (16, 16.N) by suitable clocking of the bridge circuit between the DC-side capacitance (18, 18.N) and the sub-grid (10).

2. The method according to claim 1, wherein, when determining the compensation voltage and / or the compensation current (Komp), different harmonics of the harmonic distortion current at different multiples of the grid frequency (f0) can be taken into account independently of one another.

3. Method according to claim 2, wherein a respective harmonic controller (24.3, 24.5, 24.n) uses the mains voltage (Uac) to determine a respective harmonic contribution (Uist3te Harm ), and using the respective harmonic contribution Harm) as a controlled variable, a respective harmonic compensation contribution to the compensation voltage (Komp) is determined.

4. Method according to one of the preceding claims, wherein a DC voltage (UDC, UDC.N) is detected, which is applied to the capacitor (18, 18.N) connected on the DC side of the power converter (16, 16.N), and wherein a DC control (26, 30) determines a DC control contribution to the compensation voltage and / or to the compensation current 22-438-P-WO - 23 - filed version (Comp) using the DC voltage (UDC, UDC.N) such that the DC voltage (UDC, UDC.N) is regulated to a predeterminable level by suitable active power exchange via the bridge circuit.

5. The method according to claim 4, wherein the DC control (26) comprises an f(P)-PI controller which receives as input value a difference (∆I d ) between an active power current setpoint (I d soll ) and an actual active power current value (I d ist) of the active power exchanged via the bridge circuit, whereby the active power current setpoint (I d soll ) depending on a difference between an actual value (U d is c t ) of the DC voltage (UDC, UDC.N) and the specified level (U d So c ll ) of the DC voltage (UDC, UDC.N) is determined, and wherein the active power exchange via the bridge circuit is controlled by changing the AC-side frequency ( ^^^^^^^^ ^^^^ ^^^^ ^^^^) of the power converter (16, 16.N) as a function of the output value of the f(P)-PI controller.

6. The method according to claim 4, wherein the DC control (30) comprises a P controller (P) which receives as input value a difference between an actual value (U d is c t ) of the DC voltage (UDC, UDC.N) and the specified level (U d So c ll) of the DC voltage (UDC, UDC.N), and the active power exchange via the bridge circuit is controlled by specifying a current setpoint (I WRdqsoll ) is controlled as a function of the output value of the P-controller.

7. The method according to claim 4 or 5, wherein the electrical load current (I_Last) comprises a reactive power current (I_q) at the grid frequency (f0) and a reactive power controller (28) determines a reactive power control contribution to the compensation voltage (Komp), wherein the reactive power controller (28) comprises a U(Q)-PI controller which receives as input value a difference (∆ ^^^^ ^^^^ ) between a reactive power current setpoint ( ^^^^ ^ ^ ^ ^ ^ ^ ^^ ^^^^ ^^^^ ^^^^) and a reactive power current actual value ( ^^^^ ^^^^^^^^ ^^^^ ^^^^) of the reactive power exchanged via the bridge circuit, whereby the reactive power current setpoint ( ^^^^ ^^ ^^ ^ ^^^ ^^^^ ^^^^ ^^^^) for the reactive power depending on a difference between a setpoint and an actual value of the mains voltage (� ^^^^ ^ ^ ^ ^ ^ ^^ ^ ^^ ^ ^ ^ ^ ^ ^ ^ ^^^ �) is specified such that the reactive power exchange at the connection point (AP) is reduced.

8. Method according to one of the preceding claims, wherein the method comprises the further steps of: - detecting the load current (I_Load) and the grid current (I_Grid), - determining a grid current setpoint ( ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^) using the load current (I_Last), 22-438-P-WO - 24 - submitted version - determining a target output voltage ( ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^) of the power converter as a function of the difference between an actual mains current value ( ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^) and the mains current target value ( ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^), - generating a control signal (PWMS, PWMS.N) from the determined target output voltage ( ^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^) and clocking the bridge circuit as a function of the control signal (PWMS, PWMS.N) in order to generate the compensation current (Komp).

9. The method according to claim 8, wherein determining the mains current setpoint (^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^) comprises bandpass filtering of the load current (I_Last), wherein the center frequency of the bandpass depends on the mains frequency (f0).

10. The method according to one of the preceding claims, wherein the load current (I_Last), the output current (I_filter) of the power converter (16, 16.N) and / or the mains voltage (U_ac) are detected and used to pre-control the output voltage (U_dq) of the power converter (16, 16.N).

11. Control unit (20, 20.N) for reducing harmonic power flows via a connection point (AP) of a sub-grid (10) to a higher-level AC supply grid (12), wherein a grid current (I_grid) flows between the sub-grid (10) and the higher-level AC supply grid (12) via the connection point (AP), wherein the sub-grid (10) has at least one load (14) which draws a load current (I_load) from the sub-grid (10), which load current comprises an active power current (I_d) at a grid frequency (f0) and a harmonic distortion current at one or more integer multiples of the grid frequency (f0), wherein the sub-grid (10) further has a power converter (16, 16.N) which, by means of a bridge circuit, converts electrical power between a capacitor (18, 18.N) and the sub-network (10) connected on its AC side, wherein the control unit (20, 20.N) is designed - a mains voltage (U. ac ) of the sub-network (10), - a compensation voltage and / or a compensation current (Komp) using the mains voltage (U ac ), wherein the compensation voltage and / or the compensation current (Komp) is provided for generation by the power converter (16, 16.N) and is suitable for reducing the harmonic distortion current at at least a multiple of the mains frequency (f0) in the mains current (I_Netz), and - outputting a control signal (PWMS, PWMS.N) to the power converter (16, 16.N), wherein the compensation voltage and / or the compensation current (Komp) is generated by the power converter (16, 16.N) using the control signal (PWMS, PWMS.N) 22-438-P-WO - 25 - filed version can be generated by suitable timing of the bridge circuit between the DC-side capacitance (18, 18.N) and the sub-network (10).

12. Control unit according to claim 11, wherein the control unit (20, 20.N) is designed to generate a respective harmonic compensation contribution to the compensation voltage (Komp) for different harmonic contributions (U3 is t t e Harm , U5 is t t e Harm , Harm ) at different multiples of the mains frequency (f0) independently of one another.

13. Control unit according to claim 12, wherein the control unit (20, 20.N) comprises respective harmonic controllers (24.3, 24.5, 24.n), by means of which, using the mains voltage (U ac ) a respective harmonic contribution (U3 is t t e Harm , U5 is t t e Harm , U n is t t e Harm) can be determined, whereby using the respective harmonic contribution U istnte Harm ) as a controlled variable, the respective harmonic compensation contribution to the compensation voltage (Komp) can be determined.

14. Control unit according to one of claims 11 to 13, wherein the control unit (20, 20.N) is designed to receive a DC voltage (UDC, UDC.N) which is applied to the capacitor (18, 18.N) connected on the DC side of the power converter (16, 16.N), and wherein the control unit comprises a DC controller (26, 30) by means of which a DC controller contribution to the compensation voltage and / or to the compensation current (Komp) can be determined using the DC voltage (UDC, UDC.N) such that the DC voltage (UDC, UDC.N) is regulated to a predeterminable level by suitable active power exchange via the bridge circuit.

15. Control unit according to one of claims 11 to 14, wherein the load current (I_Last) has a reactive power current (I_q) at the mains frequency (f0) and wherein the control unit (20, 20.N) is designed to carry out reactive power control (28), by means of which a reactive power control contribution to the compensation voltage (Komp) can be determined such that the reactive power exchange at the connection point (AP) is reduced.

16. Sub-network with a control unit (20, 20.N) according to one of claims 11 to 15, wherein the sub-network (10) is connected to the higher-level AC supply network at the connection point (AP), wherein the sub-network has the at least one load (14) which is designed to draw the load current (I_Last) from the sub-network (10), which load current comprises the active power current (I_d) at a network frequency (f0) and the harmonic distortion current at one or more integer multiples of the network frequency (f0), wherein the sub-network further has the power converter (16, 16.N), which is designed to convert electrical power between the on its DC side by means of the bridge circuit. 22-438-P-WO - 26 - filed version connected capacitance (18, 18.N) and the sub-grid (10) connected on its AC side in such a way that harmonic power flows via the connection point (AP) are reduced.

17. Sub-grid according to claim 16, wherein the sub-grid (10) comprises a plurality of power converters (16, 16.N), wherein the various power converters (16, 16.N) are each configured to generate a compensation voltage and / or a compensation current (Komp) for reducing respectively different harmonic contributions (Uist ist3te Harm , U5te Harm , U istnte Harm ) at different multiples of the grid frequency (f0).

18. Sub-grid according to claim 17, wherein the power converters (16, 16.N) are connected to a higher-level control unit and the higher-level control unit outputs a respective control signal (PWMS, PWMS.N) to a respective one of the plurality of power converters (16, 16.N), wherein the respective control signal (PWMS, PWMS.N) is set up to generate the compensation voltage and / or the compensation current (Komp) by the respective power converter (PWMS, PWMS.N) to reduce the harmonic distortion current at one or more multiples of the grid frequency (f0) and / or to reduce the reactive power exchange at the connection point (AP).

19. Subnetwork according to one of claims 16 to 18, wherein the subnetwork is galvanically isolated from the higher-level AC supply network, in particular by a transformer at the connection point. 20.Sub-network according to one of claims 16 to 19, wherein the power converter is connected to the sub-network without galvanic isolation.

Images