DE102012222948A1 - Method for stabilizing AC mains in low-power island net i.e. ship net, at e.g. mining industry, involves determining target idling frequency and target idling main voltage from actual active power - Google Patents

Abstract

The method involves detecting an actual active power and an actual reactive power during power exchange between a net inverter and AC mains. Gains of a switching command for the net inverter, the actual active power and a target idling frequency are determined. The net inverter is controlled based on the switching command. The target idling frequency and target idling main voltage are determined from the actual active power. Rules of direct voltage of an intermediate circuit are determined by buffering energy in the actuator. An independent claim is also included for a circuit device for stabilizing AC mains.

Description

The present invention relates to a method for stabilizing an AC network and a corresponding circuit arrangement.

In low-performance isolated networks and possibly also in a so-called "SmartGrid interconnected network", there is sometimes insufficient effective and reactive generator power available for the dynamically stable operation of large, inverter-fed consumers. As a result, voltage harmonics or network breakdowns can occur in the network, which impair the security of supply or lead to damage.

• – Auf der einen Seite sinkt die Zahl der leistungsfähigen Kraftwerke mit der Fähigkeit zur Regelung von Netzfrequenz und Netzspannung. Die Menge der rotierenden Schwungmassen (Generatoren, Turbinen), die durch ihre Trägheit kurzzeitige Schwankungen der benötigten Verbraucherleistung puffern kann, wird kleiner. Stattdessen nimmt die Zahl der erneuerbaren Energiequellen zu, die über Stromrichter – in erster Näherung unabhängig von Betriebspunktschwankungen im Netz – ihre Energie einspeisen und sich somit nicht an der primären Frequenzregelung und nicht (oder nur teilweise) an der Spannungsregelung im Netz beteiligen.
• – Auf der anderen Seite steigt bei den elektrischen Verbrauchern die Zahl der Konstant-Leistungssenken, also Verbrauchern, die mithilfe von leistungselektronischen Schaltungen die momentan benötigte Leistung dem Netz entnehmen – in erster Näherung unabhängig von Frequenz- und Spannungsschwankungen des Netzes. Beispielsweise ist es aus Energieeffizienzgründen häufig sinnvoll, bislang direkt am Netz angeschlossene elektrische Antriebe (typisch Pumpen, Lüfter) mit geregelten Wechselrichtern im optimalen Betriebspunkt zu betreiben. Teilweise wird dies bereits auch vom Gesetzgeber gefordert. Durch die leistungselektronische Schaltung wird der Leistungsbedarf des Antriebs allerdings von der Netzfrequenz entkoppelt, abhängig von der Ausprägung des Wechselrichters gilt dies auch für die Netzspannung. Dies kann z. B. für eine optimale Prozessführung zwingend notwendig sein. In vielen Anwendungsfällen wäre diese Entkopplung jedoch nicht erforderlich.

The development in the area of electrical networks is characterized by several trends:

• - On the one hand, the number of powerful power plants with the ability to control grid frequency and grid voltage is decreasing. The amount of rotating flywheels (generators, turbines), which can buffer by their inertia short-term fluctuations in the required consumer power, is smaller. Instead, there is an increase in the number of renewable energy sources that use power converters - in a first approximation, independent of operating point fluctuations in the grid - to feed their energy and thus not participate in the primary frequency control and not (or only partially) in the voltage regulation in the grid.
• - On the other hand, the number of constant-power sinks, ie consumers, which use power-electronic circuits to extract the power currently required from the grid - in the first approximation, is independent of frequency and voltage fluctuations in the network. For reasons of energy efficiency, for example, it is often sensible to operate electrical drives (typically pumps, fans) with regulated inverters that were previously connected directly to the grid at the optimum operating point. In part, this is already required by the legislature. However, due to the power electronic circuit, the power requirement of the drive is decoupled from the mains frequency; depending on the characteristics of the inverter, this also applies to the mains voltage. This can be z. B. mandatory for optimal process control. In many applications, however, this decoupling would not be required.

This can be used to derive two opposing developments, each of which affects grid stability: In comparison to the required consumer power, the available control power for short-term power fluctuations decreases. At the same time, the number of consumers is gaining power from the grid regardless of frequency and voltage fluctuations. As a result of the conversion of electrical energy generation, it is expected that the grid stability for the future interconnected grid will be impaired in the future. Reduced stability is already a problem for the increasing number of island networks. For example, large electric drives in ship nets should work stably under partial load operation even with minimal generator power since fuel can be saved by switching off generators.

Instability of electrical networks is u. a. characterized by the occurrence of significant harmonics in the mains voltage, which can lead to the damage of system parts, as well as by violating the permissible voltage frequency and voltage amplitude tolerance limits, which leads to power failure or to power off. The availability of sufficient control power in the grid is therefore crucial for the security of supply.

Particularly in isolated networks, strongly varying output is available. Switching generator and consumer loads on and off can significantly change the network impedance. In particular, active grid feeds often comprise passive filter circuits (eg L-C-L filters) in order to reduce the switching-frequency components of the inverter output voltages to standard-compliant dimensions. The resonance overshoot of the filter transfer function usually increases with the network impedance, in addition, the resonance frequency can shift significantly. This can lead to instabilities in the current loop of the inverter because the controller design is for nominal operating conditions. As a remedy are conceivable, for example elaborate parameter adaptations or an improvement in controller performance (e.g., smaller sample time, smaller dead times, etc.). As a result, additional costs for increased development and commissioning costs are expected, as well as additional costs for increased hardware expenditure (for example for computing power).

So far, to ensure a stable network operation, for example, a sufficiently large control power for the primary control of active and reactive power has been kept in the network. Alternatively, a dynamic support of the network was made exclusively with reactive power depending on the mains voltage deviation, or there was a shutdown and connection of generators and loads in a secondary control as an additional measure. Likewise, additional active filters can be used to dampen resonance effects. Furthermore, grid-friendly drives so far characterized by the fact that they take the network power sinusoidal currents and an effective factor cosφ = 1 (ie currents and voltages are in phase). This serves the efficiency but only slightly the stability.

From the publication EP 1 990 906 B1 For example, a power converter is known, in particular for a watercraft, which has a supply network, a supply bus and a plurality of drive units. Several such power converter power converters are connected to the common supply bus, and a voltage request signal indicating the voltage to be achieved at the power terminals of each power converter is provided by a power management system.

The object of the present invention is to be able to better stabilize an AC network with one or more consumers.

• – aktives Wechselrichten eines Wechselstroms des Wechselstromnetzes in einen Gleichstrom eines Zwischenkreises durch einen Netzwechselrichter,
• – Erfassen einer Ist-Wirkleistung und einer Ist-Blindleistung eines Leistungsaustausches zwischen dem Netzwechselrichter und dem Wechselstromnetz,
• – Gewinnen eines Schaltbefehls für den Netzwechselrichter in Abhängigkeit von der Ist-Wirkleistung und einer Soll-Leerlaufnetzfrequenz sowie in Abhängigkeit von der Ist-Blindleistung und einer Soll-Leerlaufnetzspannung,
• – Steuern oder Regeln des Netzwechselrichters anhand des Schaltbefehls,
• – Ermitteln der Soll-Leerlaufnetzfrequenz und der Soll-Leerlaufnetzspannung aus der Ist-Wirkleistung, einer vorgegebenen verfügbaren Netzleistung des Wechselstromnetzes, einer Ist-Verbraucherleistung eines an den Zwischenkreis angeschlossenen Stellglieds und einer vorgegebenen Soll-Verbraucherleistung für das Stellglied, und
• – Regeln einer Gleichspannung des Zwischenkreises mittels Energiepufferung in dem Stellglied.

According to the invention, this object is achieved by a method for stabilizing an AC network

• Active switching of an alternating current of the alternating current network into a direct current of an intermediate circuit by a network inverter,
• Detecting an actual active power and an actual reactive power of a power exchange between the grid inverter and the AC grid,
• Obtaining a switching command for the network inverter as a function of the actual active power and a nominal idling network frequency and as a function of the actual reactive power and a nominal open circuit voltage,
• - controlling or regulating the network inverter by means of the switching command,
• Determining the nominal idling network frequency and the nominal neutral mains voltage from the actual active power, a predetermined available network power of the AC mains, an actual consumer power of an actuator connected to the intermediate circuit and a predetermined desired consumer power for the actuator, and
• - Controlling a DC voltage of the DC link by means of energy buffering in the actuator.

• – einen Netzwechselrichter zum Umrichten eines Wechselstroms des Wechselstromnetzes in einen aktiven Wechselrichten eines Zwischenkreises,
• – ein an den Zwischenkreis angeschlossenes Stellglied mit Energiepuffer,
• – eine Leistungsmesseinrichtung zum Erfassen einer Ist-Wirkleistung und einer Ist-Blindleistung eines Leistungsaustausches zwischen dem Netzwechselrichter und dem Wechselstromnetz,
• – eine Ansteuereinrichtung zum Gewinnen eines Schaltbefehls für den Netzwechselrichter in Abhängigkeit von der Ist-Wirkleistung und einer Soll-Leerlaufnetzfrequenz sowie in Abhängigkeit von der Ist-Blindleistung und einer Soll-Leerlaufnetzspannung, und zum Steuern oder Regeln des Netzwechselrichters anhand des Schaltbefehls,
• – eine Ermittlungseinrichtung zum Ermitteln der Soll-Leerlaufnetzfrequenz und der Soll-Leerlaufnetzspannung aus der Ist-Wirkleistung, einer vorgegebenen verfügbaren Netzleistung des Wechselstromnetzes, einer Ist-Verbraucherleistung des Stellglieds und einer vorgegebenen Soll-Verbraucherleistung für das Stellglied,
• – einer Regelungseinrichtung zum Regeln einer Gleichspannung des Zwischenkreises mittels des Energiepuffers des Stellglieds.

In addition, the invention provides a circuit arrangement for stabilizing an AC network comprising

• A network inverter for converting an alternating current of the alternating current network into an active inversion of an intermediate circuit,
• An actuator with an energy buffer connected to the DC link,
• A power measuring device for detecting an actual active power and an actual reactive power of a power exchange between the mains inverter and the AC mains,
• A control device for obtaining a switching command for the network inverter as a function of the actual active power and a nominal idling network frequency and as a function of the actual reactive power and a nominal open circuit voltage, and for controlling or regulating the network inverter on the basis of the switching command,
• A determination device for determining the nominal idle network frequency and the nominal neutral network voltage from the actual active power, a predetermined available network power of the AC network, an actual consumer power of the actuator and a predetermined desired consumer power for the actuator,
• - A control device for controlling a DC voltage of the intermediate circuit by means of the energy buffer of the actuator.

Advantageously, therefore, energy is buffered in an actuator, and the energy is used to adjust the frequency or voltage of a network inverter and thus to stabilize the AC network in terms of frequency and voltage.

In one embodiment, the determination of the nominal network frequency takes place on the basis of a predefined first table from the actual active power and the nominal idling network frequency and a subsequent temporal smoothing. Analogously to this, the determination of the setpoint network voltage can take place on the basis of a predefined second table from the actual reactive power and the nominal open circuit voltage and a subsequent time smoothing. Through the respective tables, the setpoints can be determined very easily. The smoothing of the time also ensures that a stable operating point can be achieved despite fast dynamic processes in the network.

Preferably, the actuator comprises a drive and a drive inverter, with which a direct current of the intermediate circuit is converted into an alternating current for the drive. Thus, network-operated drives for network stabilization can be used.

In particular, the energy buffering can take place in a load moved by the drive. For example, the kinetic energy of a load can be recovered to stabilize the network.

According to a further development, a switching command for the drive inverter is determined from an actual current from the drive inverter to the drive and a predetermined desired magnetic flux of the drive. This is possible Realize a power take-off via the drive inverter specifically for the respective drive via the setpoint flux as the specified motor size.

The switching command for the drive inverter can also be determined by means of a torque-flow control using a predetermined setpoint torque. The regulation thus allows the energy in the DC link to be adjusted as required.

The setpoint torque can be obtained from an actual intermediate circuit voltage and a predetermined setpoint intermediate circuit voltage by regulation. This makes it possible to track the setpoint torque directly to the intermediate circuit voltage.

Furthermore, the actual active power of the network inverter can be used to control the setpoint torque. Advantageously, a pre-control of the load power can thus be achieved by the power actual value known from the network-side inverter.

The abovementioned method features also represent optional functional features of the circuit arrangement according to the invention.

The present invention will now be explained in more detail with reference to the accompanying drawings, in which:

1 an active power statistic characteristic for the frequency;

2 an active power statics characteristic for the voltage;

3 two active power statics characteristic curves for determining a common operating point of several generators in the network;

4 a block diagram of the method according to the invention for power frequency and mains voltage stabilization with drive inverters;

5 a block diagram for obtaining the phase angle φ of the output voltage and

6 an alternative block diagram to that of 5 ,

The embodiments described in more detail below represent preferred embodiments of the present invention.

An inventive control and control method is suitable for inverter-powered consumers. The method is characterized by the fact that the grid-side inverter does not regulate the power flow between the grid and the DC link in accordance with a power setpoint (prior art). Instead, the task of the network-side inverter is to counteract frequency changes and voltage changes in the network and thus to stabilize the network, in particular, dynamic processes that are faster than a network period, can be stabilized. For slow load swings in the network, a system according to the prior art would be sufficient. An actuator as a load on the DC link or a load-side inverter regulates the DC link and influences suitably the power consumption of the load.

A higher-level power management adapts the frequency and voltage setpoints for the grid-side inverter so that a desired power exchange between load and grid is realized on average. For this, e.g. a mean desired torque or a mean target speed for the load can be specified if enough generator power is available in the network.

The invention is based on the following findings: The load balancing between the generator units in the network, which is necessary for stable operation at an operating point with a common frequency and amplitude, preferably takes place by means of so-called static characteristics (cf. 1 to 3 ). The statics may include smoothing elements (PT1) to allow adaptation to the present time constants of the respective network. For example, the frequency of a generator decreases with increasing power output. It results, for example, in 1 illustrated active power statics characteristic fL for idling (generator E are characterized by a negative, consumer V by a positive active power.). A corresponding relationship between reactive power and voltage is due to the reactive power statistic UL for idling 2 given.

Each power generator in the grid, which actively contributes to frequency and voltage maintenance, uses these characteristics to determine its current setpoint values for output voltage and frequency. This is often referred to as primary regulation. For example, if all generating units have the same nominal power and characteristics, then there will be an F-U operating point in the grid at which all generators contribute the same power to the required total load. It thus results in a symmetrical load distribution without further communication between the producers with each other.

The operation of such a system is a simple example of the Active power characteristic with two generators and a load load Pv explained (cf. 3 ). For the output frequency of the first and second generators, k1 and k2 have slopes and the idling frequency f0 selected as characteristic curves: fE1 = f0 + k1 * P1, fE2 = f0 + k2 * P2.

With the condition for the power sum P1 + P2 + PV = 0, the frequency of the second generator is subject to the following dependence on the power of the first generator: fE2 = f0 + k2 · (-P1 - PV).

Therefore, the characteristic of the second generator can be mirrored and shifted by PV in the P-f diagram of the first generator to illustrate the resulting operating point with the common frequency fA. The first producer assumes the share PE1 of the necessary total output.

The primary control in networks is typically a fast P-control, which can not balance out deviations. Stationary accuracy is achieved by superimposed slow secondary controls, the z. B. shift the idle points of the static characteristics so that the net nominal frequency is reached on average. For this purpose, the manual connection of additional producers may be necessary.

Constant power sinks have specific effects on a grid. The destabilizing effect of regular regulated drive inverters on a grid shows a simple consideration: Let's assume a temporarily decreasing mains voltage. The power control of the inverter then increases the active current to keep the power constant. The additional current causes additional voltage drops at the network impedances and increases the losses in the network. Consequently, the mains voltage drops in addition and the power generators are additionally loaded.

Mains frequency changes have no direct effect on the power requirement of a regular regulated drive inverter. However, the network will be e.g. with falling line frequency also not stabilized by reduced power consumption (according to a P-f characteristic curve). Indirectly, the total power demand in the network may even increase in this example as the frequency-dependent impedance of the grid inductances decreases. Thus, the reactive current increases, and the losses of the resistive components of the network impedance increase.

The following is a concrete example based on 4 shown. It shows a consumer 1 that has an active power inverter 2 from the network 3 is fed. In the present example, a 3-phase network is shown. But it can also be a network with only one or two phases or the like. In the present example, there is also between the grid and the grid inverter 2 a standard network filter 4 connected.

On the DC side of the network inverter 2 there is a DC link 5 with a DC link capacitor CZK. At this the DC link voltage Vdc is applied.

The DC link is used to supply the load 1 , which can be here generally an actuator with energy buffering and regulation of the intermediate circuit voltage Vdc. In the present case, the consumer has a drive inverter 6 on, the DC of the DC link 5 in an alternating current (here also 3-phase system) for a drive or motor 7 converts. The motor 7 for example, turns a load 8th with an actual speed NIst. The load may be, for example, a propeller or a pump. The drive inverter 6 is part of a regulation of the intermediate circuit voltage Vdc described in more detail below.

In the power inverter 2 it is an active grid inverter, in the power flow from both the AC grid in the DC link 5 as well as vice versa is possible and sinusoidal alternating currents (with adjustable active and reactive component) in the network 3 can be impressed (see eg international Standard IEC / TS 62578 : "Power electronics system and equipment - Operation conditions and characteristics of active infeed converter applications").

A typical arrangement according to the prior art is a self-commutated voltage source inverter with IGBTs in the B6 bridge circuit, which is clocked with a pulse frequency in the kHz range and realize on average via a PWM clock voltage space vector with a defined amplitude and defined phase angle at the output of the inverter can. It is customary to use a current control to calculate the voltage room pointer in such a way that the desired active and reactive current at the connection point of the network inverter results. With the degrees of freedom amplitude and angle (or active and reactive current), such an inverter is in principle able to provide a voltage with a defined amplitude and frequency at its terminals.

According to the prior art, therefore, the exchanged with the network Active power is controlled so that a required performance required by the process is achieved (for example due to the current power requirement of a drive connected to the DC link or for recharging the DC link capacitance (see eg EP 1 990 906 B1 )).

Regardless of the effective axis, the reactive current can be regulated to a setpoint. Due to the voltage drop at the mains inductances, the amount of mains voltage can be influenced by the reactive current. This is already being used today and e.g. in grid connection standards for power generation plants (see Bundesverband der Energie- und Wasserwirtschaft e.V .: "Guideline for Connection and Parallel Operation of Generation Plants on the Medium-Voltage Grid", 2008). Reactive power generated by converters thus serves as network support or stabilization. In the foreground stands the voltage maintenance or the compensation of the voltage drops at the network impedances with load changes.

According to the present invention, the effective axis is now also included in the network stabilization, which is needed for frequency maintenance in the network. For short-term dynamic processes, the consumer thus acts in the network with its full power as an additional generator unit. Overall, this provides a concrete method by which converter loads in the network with their entire dynamic power contribute to primary regulation of voltage and frequency (or reactive power and active power) and thus to the dynamic stabilization of the network.

For this purpose, with the grid inverter 2 instead of the active power, the output frequency of the line rectifier is regulated to a defined setpoint. In particular is - for example, with a smoothing member 9 . 10 - ensures that the output frequency of the network inverter can change only slowly compared to its PDM cycle. This corresponds to a producer with generator an inertial speed change due to the flywheel. The grid inverter 2 acts in this way a frequency change in the network 3 stabilizing against. If the mains frequency changes at initially constant output frequency of the network inverter 2 , so results in a phase angle between line inverter voltage and line voltage, which leads to a power consumption or power output depending on the sign. This power change accordingly leads to an energy increase or decrease in the DC link.

The load-side consumer 1 - In the example of the drive inverter 6 with electric drive 7 and mechanical load 8th - this energy change must be in the DC link 5 compensate by changing the power accordingly. Advantageously, this is an intermediate circuit voltage regulator 11 used, which is the set torque MSoll of the engine 7 so adapts that the DC link 5 a power is taken, which counteracts the voltage change. The drive power is accordingly subject to temporary fluctuations, for which the application in question must be suitable. An example would be an optional auxiliary drive on ships. Stationary accuracy can (similar to grid regulation) by a superimposed slow secondary control 12 can be achieved, the idling frequency f0 (see. 3 ) of a network-side frequency statics 13 if the condition of the network or additional generator power allows it.

As well as the frequency, the amplitude of the line-side output voltage of the network inverter is 2 only slowly changed and thus affects voltage changes in the network 3 opposite. Because the voltage amplitude is on the mains 3 exchanged reactive power is the load-side load 1 not affected by this stabilization.

The example of 4 will now be described in detail: The on the load side inverter 6 connected electrical load does not have to be a drive. In general, for the load side, it is only required that the converter or, in general, the actuator for regulating the intermediate circuit voltage Vdc be able to operate as described above. This can be achieved by controlling the consumed power. In principle, a load resistor would be conceivable as an electrical load.

If necessary, the application must first be brought into a suitable operating state / load state for the possibility of active power support. For stabilization, the power must be increased and decreased. In the case of a load resistance (chopper or braking resistor), an operating point with power consumption must first be assumed before a corresponding discharge is possible.

For controlling the drive inverter 6 here stands a torque and flow control 14 to disposal. It generates from the AC currents IRST (currents of the output-side phases R, S and T of the drive inverter 6 ), a target magnetic flux .phi.Soll and the target torque MSoll switching commands SB for the drive inverter 6 , It exploits the fact that the product of magnetic flux and current leads to the respective torque. The target torque MSoll in turn is from the Vdc control 11 as a function of the intermediate circuit voltage Vdc and a desired intermediate circuit voltage Vdc, setpoint and an actual active power regulated. The actual active power PIst comes from a performance measuring device 15 that the power flow in the grid inverter 2 or determined on the basis of the grid sizes IUVW and VUVW. The setpoint values Φ setpoint and Vdc, setpoint for the motor flux and the DC link voltage can be determined by a higher-level control (in 4 not shown) suitable for the respective application and, for example, given constant. Depending on the type of machine and the speed, it may be necessary to lower the flux or to increase the intermediate circuit voltage Vdc, but this is not important for stabilization. Depending on the type of line rectifier (eg with step-up converter with IGBT in B6 circuit), the requirement that the condition Vdc>"Amplitude of the line-to-line voltage" must be met must also be met.

The mains side is connected to the power measuring device 15 In addition to the actual active power PIst and the actual reactive power QIst determined. Using the active power frequency statics 13 and the smoothing member connected thereto 9 is determined from the actual active power PIst and a nominal idling network frequency fNL, setpoint nominal network frequency fN, target. Analog is with a reactive power voltage statics 16 (see. 2 ) and the trailing smoothing member 10 from the actual reactive power QIst and a nominal open circuit voltage VNL, setpoint to determine a nominal mains voltage VN, nominal. From the nominal mains frequency fN, nominal and the nominal mains voltage VN, setpoint generates a frequency-voltage control 17 Switching commands SB for the grid inverter 2 , So can the network inverter 2 known power actual value (PIst and Qist) can be used advantageously for the precontrol of the load power. This reduces the deviations of the intermediate circuit voltage Vdc from the setpoint.

The blocks real power statics 13 and reactive power statics 16 correspond with the 1 and 2 , In the waveform from the controller to the switching commands is a smoothing (smoothing 9 and 10 ), which ensures a temporary constancy of the output frequency and output voltage of the grid inverter and is thus decisive for the degree of grid stabilization.

For the "regulation" of the mains frequency can in the simplest case in the block 17 an integrator (see integrator 18 in the 5 and 6 ) be used. The integrator uses the frequency to calculate the current phase angle φ of the output voltage. For the "regulation" of the mains voltage amplitude by block 17 can a Proprotionalitätsglied (see Proportionalitätsglied 19 from 5 ) be used. By angle and amplitude, the switching commands SB - for example, by a lookup table - clearly determined in a tax rate.

A complex, but in principle possible alternative for the integrator and the proportional link in the block 17 would be a fast current control, the network currents just impressed in such a way that the output voltage assumes the desired frequency and amplitude. As an alternative to reactive power / active power, the reactive current / active current can also be used.

In the higher-level control 12 (Performance management) are the setpoints fNL, setpoint and VNL, setpoint for the statics 13 and 16 depending on the desired behavior of the overall application, in particular depending on the actual rotational speed NIst, the actual torque MIst, an externally specified target rotational speed NSoll, an externally specified target torque MSoll and a likewise externally predetermined available network power PN generated. For example, a certain drive power in a stand-alone grid can only be provided if, on average, sufficient generator power is available.

With slow control (e.g., PI), the static idle points can be tracked to achieve on average steady state accuracy for the torque of a drive or the network amplitude. Comparable effect on power plants would have the so-called secondary regulation.

In principle, the deviation between actual speed value and setpoint value can also be used by a PI controller in order to track the idling frequency f0. Due to the smoothing in the static channel, however, this is meaningful and stable only for drives with slow dynamics.

The use of linear characteristic curves for statics makes sense, but not mandatory. Generally, the slope should be monotone.

In principle, it would also be conceivable to set the desired drive power for the load-side converter 6 to calculate directly via a characteristic curve depending on the current grid frequency. The grid inverter 2 Then, as in the prior art, the network active power could be based on the regulation of the intermediate circuit voltage Vdc as well as regulate the reactive power. However, in this way it would only be possible with great effort to ensure that the output frequency of the network inverter 2 only slowly compared to the PWM clock changes.

Optionally, additional attenuators can be implemented to avoid power swings between the grid and the stabilizing grid inverter. In principle, this can be achieved with an additive D component (a PID controller) in the frequency control path. For example, according to 5 by doing block 17 an integrator 18 with upstream first proportional element 19 be provided to determine from the nominal mains frequency fN, set an angle φ, in which case the output of the integrator 18 with the output of a second proportionality element 20 is summed up into the same as in the first proportionality element 19 the nominal network frequency fN, target is fed.

In an alternative embodiment according to 6 is in the block 17 also an integrator 18 provided at the output of the angle φ is provided. The input signal of the integrator 18 consists of a sum of the nominal mains frequency fN, setpoint and an output signal of a differentiator 21 , The input signal of the differentiator 21 is by means of a proportionality element 22 from the nominal grid frequency fN, Soll won.

For the principal function of "stabilizing the network" damping (for example with a D-component) is not mandatory. However, in most cases such damping will be used as performance is generally improved.

The proposed approach differs from methods that use modeling to try to replicate the performance of synchronous generators with a drive as closely as possible. On the one hand, the present approach is much more practicable since there is hardly any effort in comparison to set model parameters. On the other hand, the synchronous machine models are intended for generating plants with line rectifiers, which should simulate the network behavior of synchronous generators as closely as possible. In the present case, however, consumer loads, which together require a significant proportion of the producer's output, are expected to contribute to grid stabilization.

According to the invention, the operation of large drive converters is thus possible even on very low-power networks. This is of particular interest e.g. in ship nets, where there is a requirement to operate large traction drives in partial load with minimal generator configuration (fuel economy). In this case, resonance effects in the mains voltage as well as mains voltage and frequency fluctuations are effectively suppressed. The network is thus stabilized. Similar requirements are to be expected in other large-scale stand-alone networks, e.g. in mining.

In addition, in the so-called "SmartGrid networks" with a high proportion of decentralized generation with inverter feeds, the question of dynamic grid stability can become considerably more important. Since the security of supply can be affected on a large scale, there is a corresponding economic importance.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 1990906 B1 [0008, 0044]

Cited non-patent literature

• Standard IEC / TS 62578 [0042]

Claims (10)

Method for stabilizing an AC network ( 3 ), characterized by - active switching of an alternating current of the alternating current network ( 3 ) in a DC current of a DC link ( 5 ) by a network inverter ( 2 ), - detecting an actual active power (PIst) and an actual reactive power (QIst) of a power exchange between the power inverter ( 2 ) and the AC mains ( 3 ), - obtaining a switching command (SB) for the network inverter ( 2 ) as a function of the actual active power (PIst) and a nominal open-circuit network frequency (fNL, setpoint) and as a function of the actual reactive power (QIst) and a nominal open circuit voltage (VNL, setpoint), - controlling or regulating the network inverter ( 2 ) based on the switching command (SB), - Determining the nominal idling network frequency (fNL, Soll) and the target neutral line voltage (VNL, Soll) from the actual active power (PIst), a predetermined available network power (PN) of the AC network ( 3 ), an actual consumer power (NIst, MIst) one to the DC link ( 5 ) connected actuator ( 1 ) and a predetermined desired consumer power (NSoll, MSoll) for the actuator, and - regulating a DC voltage (Vdc) of the intermediate circuit ( 5 ) by means of energy buffering in the actuator ( 1 ). The method of claim 1, wherein obtaining the switching command (SB) based on a predetermined first table ( 13 ) from the actual active power (PIst) and the nominal open-circuit network frequency (fNL, setpoint) and a subsequent temporal smoothing ( 9 ) he follows. Method according to claim 1 or 2, wherein the obtaining of the switching command (SB) is based on a predetermined second table ( 16 ) from the actual active power (PIst) and the nominal open circuit voltage (VNL, setpoint) and a subsequent time smoothing ( 10 ) he follows. Method according to one of the preceding claims, wherein the actuator ( 1 ) a drive ( 7 ) and a drive inverter ( 6 ), with which a direct current of the intermediate circuit ( 5 ) is converted into an alternating current for the drive. Method according to claim 4, wherein the energy buffering in one of the drive ( 7 ) moving load ( 8th ) he follows. Method according to claim 4 or 5, wherein a switching command (SB) for the drive inverter ( 6 ) from an actual current (IRst) from the drive inverter ( 6 ) to the drive ( 7 ) and a predetermined desired magnetic flux (Φ setpoint) of the drive is determined. Method according to Claim 6, in which the switching command (SB) for the drive inverter ( 6 ) by means of a torque-flow control ( 14 ) is determined based on a predefinable setpoint torque (MSoll). The method of claim 7, wherein the setpoint torque (MSoll) from an actual intermediate circuit voltage (Vdc) and a predetermined desired intermediate circuit voltage (Vdc, Soll) is obtained by regulation. Method according to claim 8, wherein the actual active power (PIst) of the network inverter ( 2 ) is used for the regulation of the setpoint torque (MSoll). Circuit arrangement for stabilizing an alternating current network ( 3 ), characterized by - a network inverter ( 2 ) for actively inverting an alternating current of the alternating current network ( 3 ) in a DC current of a DC link ( 5 ), - one to the DC link ( 5 ) connected actuator ( 1 ) with energy buffer, - a power measuring device ( 15 ) for detecting an actual active power (PIst) and an actual reactive power (QIst) of a power exchange between the grid inverter and the AC grid ( 3 ), - a control device ( 13 . 16 . 17 ) for obtaining a switching command (SB) for the network inverter ( 2 ) as a function of the actual active power (PIst) and a nominal idling network frequency (fNL, setpoint) and in dependence on the actual reactive power (QIst) and a nominal open-circuit voltage (VNL, Soll), and for controlling or regulating the network inverter ( 2 ) based on the switching command (SB), - a determination device ( 12 ) for determining the nominal open-circuit network frequency (fNL, nominal) and the nominal open-circuit network voltage (VNL, nominal) from the actual active power (PIst), a predetermined available network power (PN) of the alternating current network ( 3 ), an actual consumer power (NIst, MIst) of the actuator and a predetermined desired consumer power (NSoll, MSoll) for the actuator ( 1 ), - a control device for controlling a DC voltage (Vdc) of the intermediate circuit ( 5 ) by means of the energy buffer of the actuator ( 1 ).

Images