Multi-port grid formation control for grid interconnection

Abstract

The present application relates to a grid forming vector current control system for controlling grid interconnection, the system comprising: a first terminal configured to be connected to a first power grid, the first terminal comprising a first current control unit, a first virtual admittance unit and a first phase locked loop, PLL, unit; a second terminal configured to be connected to a second power grid, the second terminal comprising a second current control unit, a second virtual admittance unit and a second PLL unit; and a controller configured to generate a power equivalent current reference using transient power consumed by a first virtual synchronous machine to control a second virtual current source and to generate a power equivalent current reference using transient power consumed by a second virtual synchronous machine to control a first virtual current source. The present application also relates to a corresponding method.

Description

[0001] This disclosure relates to systems and methods for vector current control in grid formation—specifically for multi-port grid formation control in grid interconnection.

[0002] The rate of change in transmission and distribution grids is traditionally high. Conventional power generation using synchronous generators is increasingly being replaced by renewable energy resources connected to converters. Numerous studies have demonstrated the risk of grid instability if a utility grid is equipped with a large amount of renewable energy resources without employing any grid-supported control strategies. This instability arises because such resources do not adapt their power injection to the grid's actual power demand.

[0003] This grid-independent current / power control strategy has long been common in most converter-connected applications. However, stricter regulations have been implemented in recent decades, such as for photovoltaic (PV) inverters. Today, if the grid frequency increases above nominal levels, PV inverters must gradually reduce the injected active power. This behavior is a first step toward grid-supported power converter control, slowly adapting the injected active power to the actual needs of the grid. Such a simple approach only provides steady-state grid support; that is, it still relies on synchronous generators in the grid to handle transient load changes.

[0004] Meanwhile, more sophisticated converter control methods have been developed for isolated grids, microgrids, and rail grids. These control methods are designed to provide not only steady-state grid support but also transient grid support. Energy resources connected to converters using such control methods can completely replace conventional synchronous generators in the grid. The grid can operate without any conventional synchronous generators connected to it—a common scenario in, for example, rail grids.

[0005] To provide full grid support, it is necessary to vary the amount of power injected into the grid according to the demands of various loads connected to it. This is achieved by changing the behavior of grid-independent, constant power injection to one that more closely resembles that of a voltage source. Converters equipped with such a control method automatically adapt and share the injected power, enabling them to provide the actual power required by the various loads in the grid.

[0006] For backward compatibility with legacy equipment in AC power generation and distribution, the control method is designed to provide “inertia” to the grid, meaning that the rate of change of the grid frequency is limited.

[0007] The combined effects of voltage source characteristics and inertia result in behavior very close to that of a synchronous generator. This control method provides grid-wide support and can completely replace synchronous generators in the grid.

[0008] Generally, based on the control mechanisms of existing technologies, they can be divided into two main groups.

[0009] In the droop control approach, the control law is based on the steady-state coupling of frequency (or angle) and power flow in an inductive grid. Various implementation variations exist, and several ways to improve the approach have been proposed in recent years. In most implementation variations, the method creates a voltage reference that must be tracked by additional cascaded voltage and current control loops. In this regard, see references [1] and [3].

[0010] In the virtual generator method described, for example [2], the control law is based on the principle of simulating the behavior of a synchronous generator. The mathematical model describing the behavior of the generator is implemented directly in the controller. There are various implementation variations that differ mainly in the level of detail and the complexity of the equations implemented.

[0011] To simulate a virtual generator or virtual synchronous machine (VSM), the mathematical equations of the synchronous generator are implemented in the controller to simulate the generator's behavior. In the direct method (voltage source type), the machine equations are solved against a stator voltage given as a reference directly to the modulation stage or to subsequent cascaded voltage and current control loops. In the inverse method (current source type), the machine equations are solved against a stator current given as a reference to subsequent current control loops.

[0012] However, both methods cited above require various additional control loops to make the concepts applicable. Specifically, the cascaded voltage and current control loops provided by the direct method offer current control, including current limiting.

[0013] The current control loop and PLL track the active / reactive power reference given by the inverse method, and provide current control including current limiting. Additionally, grid synchronization and grid frequency measurements are provided.

[0014] Virtual impedance loops can simulate virtual stator impedance.

[0015] Additional components can be used to improve the performance of the virtual generator. These additional components can be at least one of the following:

[0016] Additional PLLs are used for direct methods of synchronizing with the grid before the main circuit breaker (MCB) of the closed converter or during fault conditions; automatic voltage regulators (AVRs) are used to improve voltage regulation at the PCC; and feedforward gains are used for grid current and PCC voltage to improve the performance of the PI control loop. Damper winding simulation is used to improve the damping of low-frequency oscillations. In the simplest inverse method, the virtual inertial effect is simulated using the derivative of the measured grid frequency. Furthermore, a low-pass filter is often required on the virtual impedance reference to reduce the amplification of high-frequency noise caused by the derivative used to calculate the virtual inductance.

[0017] However, advanced systems can be difficult to tune and require significant commissioning effort. Interactions may also exist between control loops due to insufficient bandwidth separation in converters with lower switching frequencies. In ordinary systems, approximately 15 parameters typically need to be tuned due to unknown interactions. In strong power grids, insufficient reference tracking performance often occurs. Furthermore, synchronization loss may occur during faults.

[0018] As discussed earlier, grid-supported control methods are becoming increasingly important in various applications—such as grid-connected battery energy storage (BESS), microgrids, PV inverters, rail interconnection, high-voltage direct current (HVDC), and uninterruptible power supplies (UPS). Transmission system operators (TSOs) in various countries are discussing changing the standards for grid-connected energy resources in such a way that grid support becomes mandatory for all grid-connected energy sources. Current grid-supported control methods are primarily based on two fundamental concepts—the "virtual generator" concept and the droop control concept. Both methods have several drawbacks, such as high complexity, difficult tuning methods with high commissioning workloads, and stability issues. Furthermore, the methods mentioned can provide sufficient control performance in both strong and weak grids simultaneously, and therefore require retuning or control mode switching during islanding events.

[0019] Besides the renewable energy sources connected by converters, HVDC lines are becoming increasingly important for transmitting large amounts of power over long distances and for coupling different power grids, such as a three-phase utility grid with a one-phase rail grid. Conventionally, such interconnectors are controlled using only constant power control or steady-state grid support. Recently, the use of virtual synchronous generator control has been proposed to provide full grid support for HVDC lines or rail interconnects.

[0020] However, advanced solutions may only provide full grid support on one side of the link, meaning full grid support cannot be provided at both terminals of the HVDC link or grid interconnection station, and there is not much energy storage in the DC link between the two terminals. However, TSOs, in particular, will prefer solutions with full grid support capabilities on both sides of the terminals.

[0021] Existing control methods for HVDC line and track interconnection can be categorized into the following solutions:

[0022] Strict current / power control on both terminals to transfer a given amount of active / reactive power from one terminal to the other. An extension of this method allows the use of the derivative (rocof) of the grid frequency to simulate inertia on one terminal (ref [4]). Typically, the other terminal is determined to control the DC link voltage and does not provide any inertia.

[0023] - Strict current / power control on one side of the terminal and grid support control on the other side. The grid support terminal may be equipped with only steady-state droop control or with full virtual synchronous generator control (see [6]). The terminal with strict power control is determined to control the DC link voltage.

[0024] - Strict current control on one side of the terminal and grid formation control on the other side. This mode is used, for example, to connect offshore wind farms to the public grid. Offshore terminal simulation provides a voltage source for the local relaxation bus (bus) of the wind farm (reference [5]). The terminal with strict power control is determined to control the DC link voltage.

[0025] - Grid support control on one or both terminals of the DC link with excessive energy storage[7].

[0026] Figure 1 An example schematic diagram of a grid interconnect / HVDC link according to prior art is shown. In this case, the DC link is represented by a capacitor bank. According to prior art, both terminals are controlled using vector current control (VCC) or an equivalent current control scheme, such that the behavior of the two converter terminals can be modeled using an equivalent current source (see [link to previous section]). Figure 2 By providing the correct setpoint through current control at the two terminals, a certain amount of power (P) can be transmitted over the link. tr Reactive power can be injected independently at both terminals. Assuming vector current control in a synchronous (dq) reference coordinate system, the correct setpoint for the current controller needs to be calculated in both d and q coordinates. Assuming steady state, i.e., each PLL reference coordinate system is aligned with the corresponding voltage at the PCC, the setpoint is given as...

[0027]

[0028]

[0029]

[0030]

[0031] For simplicity, the DC link voltage control is calculated based only on the first terminal connected to the first power grid. However, it can also be shared between the first terminal and the second terminal connected to the second power grid. Power transfer setpoint P tr It can be a quantity calculated by a slow top-level controller (secondary or tertiary control, power trading), or it can additionally contain a portion calculated by a droop curve for steady-state grid support based on a frequency measurement at one terminal. Virtual inertia can be simulated at one terminal using the frequency derivative (rocof) (see [4]). The grid frequencies at both terminals are considered to determine P. tr More complex functions are also possible.

[0032] In other applications (such as the connection of offshore wind farms to the grid), a different, more advanced control method is employed. Terminals connected to the offshore wind farm can be controlled in grid-forming mode, i.e., by simulating a voltage source (see [link to relevant documentation]). Figure 3 The exchange of active and reactive power between the second terminal and the second grid is not directly controlled. Standard vector current control is applied to the first terminal, which must ensure power balance between the first and second terminals. The setpoint for the current controller used for the first terminal is then given as...

[0033]

[0034]

[0035]

[0036] Again, through The DC link voltage is controlled by the first terminal.

[0037] In another example according to the prior art, one terminal is equipped with virtual synchronous generator control (see [link]). Figure 4The other terminal is controlled using standard vector current control. The controller on the second terminal simulates the behavior of a synchronous generator. The power exchange between the second terminal and the second grid is determined by the parameters and dynamics of the virtual synchronous generator. However, steady-state power exchange can be affected by the power setpoint entering a droop curve, which is typically implemented for the virtual generator. The setpoint for the current controller on the first terminal is again given as:

[0038]

[0039]

[0040]

[0041] Next, the DC link voltage is controlled using the first terminal. Under steady state, the active and reactive power injected into the grid by the virtual generator at the second terminal typically follows a droop curve according to the following formula.

[0042]

[0043]

[0044] The active power setpoint Pset,2 and reactive power setpoint Qset,2 are given by the slow top-level controller (secondary or tertiary control, and / or power trading).

[0045] Advanced control methods cannot provide full grid support at both ends of the interconnect / HVDC link without the excessive energy storage installed in the DC link.

[0046] One of the main concepts of this disclosure is to provide grid support / grid forming control at both terminals of the HVDC link / grid interconnect. This is achieved by using grid forming vector current control (GFVCC) and bidirectional power feedforward terms on both sides of the link. GFVCC is a control method that will be discussed in detail below.

[0047] This method is not limited to two-port systems, and it can also be extended to multi-port HVDC systems.

[0048] A unique feature of grid-forming vector current control is that it simulates the behavior of a virtual synchronous generator—a rotating wheel—connected in parallel with a current source. This rotating wheel neither consumes nor supplies any power during steady state; it only injects power and provides inertia to the system during transients. The current source, connected in parallel, acts as a fast regulator and provides the steady-state power required by the application.

[0049] According to one aspect of the invention, this control method is replicated and applied to control two terminals of an HVDC link / grid interconnect. Power injected by the virtual generator portion of the controller at the first terminal is used to create a power-equivalent current reference for a current source at the second terminal. Conversely, power injected by the virtual generator portion of the second terminal is used to create a power-equivalent current reference for a controlled current source at the first terminal. In this method, the terminals are provided with the characteristics of a virtual generator. The required power is sourced / sinked from the grid at the opposite terminal.

[0050] The amount of power grid support at each terminal, i.e., the inertial response, can be easily and independently determined using a factor—k—introduced in GFVCC from 0% to 100% of the link's nominal power. VSM Factors – for adaptation. One advantage of this scaling is that it can reduce the amount of grid support on the strong grid side without excessively interfering with the weak grid side.

[0051] In addition, the same factor—k VSM —It can be used to change the system to strict current control and reduce the virtual generator effect to zero at each terminal in the event of an independent grid failure.

[0052] Furthermore, the power flow controller can create setpoints for two current sources for the regular transmission of active / reactive power via HVDC links / grid interconnection. This power flow controller may include devices for power trading and / or steady-state grid support using droop curves or similar methods.

[0053] In addition, the DC link voltage controller can create setpoints for current sources on one or two terminals to control the DC link voltage of the link.

[0054] This disclosure relates to a grid-forming vector current control system for controlling grid interconnection. The system includes a first terminal configured to connect to a first power grid, the first terminal including a first current control unit, a first virtual admittance unit, and a first phase-locked loop (PLL) unit, wherein the first virtual admittance unit and the first PLL unit are configured to simulate the inertia of a first virtual synchronous machine (VSM), and a first virtual current source is connected in parallel to the first VSM. The system includes a second terminal configured to connect to a second power grid, the second terminal including a second current control unit, a second virtual admittance unit, and a second PLL unit, wherein the second virtual admittance unit and the second PLL unit are configured to simulate the inertia of a second virtual synchronous machine (VSM), and a second virtual current source is connected in parallel to the second VSM. The system includes a controller configured to use transient power P consumed by the first VSM. VSM1 To generate a power-equivalent current reference to control the second virtual current source, and to use the transient power P consumed by the second VSM. VSM2 To generate a power-equivalent current reference to control the first virtual current source.

[0055] Multiple implementation schemes may preferably implement the following features:

[0056] Preferably, the first terminal includes a first droop control unit, and the output signal of the first droop control unit is connected to the first virtual current source to simulate a first speed regulator.

[0057] Preferably, the second terminal includes a second droop control unit, and the output signal of the second droop control unit is connected to the second virtual current source to simulate the second speed regulator.

[0058] Preferably, the first terminal includes a first droop control unit, and the output signal of the first droop control unit is connected to the first PLL unit to simulate a first virtual mechanical speed regulator.

[0059] Preferably, the second terminal includes a second droop control unit, and the output signal of the second droop control unit is connected to the second PLL unit to simulate a second virtual mechanical speed regulator.

[0060] Preferably, the first factor K VSM1 Configured to use K VSM1 The output signal of the first virtual admittance is scaled proportionally to continuously change the characteristics of the first terminal from a voltage source to a current source.

[0061] Preferably, the second factor K VSM2 Configured to use KVSM2 The output signal of the second virtual admittance is scaled proportionally to continuously change the characteristics of the second terminal from a voltage source to a current source.

[0062] Preferably, at least one of the first droop control unit, the first virtual admittance unit, or the first PLL unit is configured to receive the factor k. VSM1 To increase or decrease the inertial response of the first terminal to interference in the first power grid.

[0063] Preferably, at least one of the second droop control unit, the second virtual admittance unit, or the second PLL unit is configured to receive the factor k. VSM2 To increase or decrease the inertial response of the second terminal to interference in the second power grid.

[0064] Preferably, K VSM1 By using K VSM1 The output signal of the first virtual admittance is scaled proportionally using K. VSM1 The scaling gain of the first PLL unit is inversely scaled and / or K is used. VSM1 The nominal power of the first VSM is scaled proportionally by scaling the first droop constant.

[0065] Preferably, K VSM2 By using K VSM2 The output signal of the second virtual admittance is scaled proportionally using K. VSM2 The proportional gain of the second PLL unit is scaled inversely and / or using K VSM2 The nominal power of the second VSM is scaled proportionally by scaling the second droop constant.

[0066] Preferably, k VSM1 Scaling of the rotation wheel factor, transient power, or inertia.

[0067] Preferably, k VSM2 Scaling of the rotation wheel factor, transient power, or inertia.

[0068] Preferably, the system further includes a third terminal configured to connect to a third power grid. The third terminal includes a third droop control unit, a third current control unit, a third virtual admittance unit, and a third PLL unit. The third virtual admittance unit and the third PLL unit are configured to simulate the inertia of a third virtual synchronous machine (VSM), and a third virtual current source is connected in parallel to the third VSM. Preferably, the controller is configured to use the transient power P consumed by the first VSM. VSM1 and the transient power P consumed by the second VSM VSM2A power-equivalent current reference is generated to control the third virtual current source, using the transient power P consumed by the first VSM. VSM1 and the transient power P consumed by the third VSM VSM3 A power-equivalent current reference is generated to control the second virtual current source, and the transient power P consumed by the second VSM is used. VSM2 and the transient power P consumed by the third VSM VSM3 A power-equivalent current reference is generated to control the first virtual current source.

[0069] This disclosure also relates to a vector current control method for controlling grid formation in grid interconnection. The method includes: providing a first terminal configured to connect to a first power grid, the first terminal including a first current control unit, a first virtual admittance unit, and a first phase-locked loop (PLL) unit; simulating the inertia of a first virtual synchronous machine (VSM) using the first virtual admittance unit and the first PLL unit; and providing a first virtual current source connected in parallel to the first VSM. The method further includes: providing a second terminal configured to connect to a second power grid, the second terminal including a second current control unit, a second virtual admittance unit, and a second PLL unit; simulating the inertia of a second virtual synchronous machine (VSM) using the second virtual admittance unit and the second PLL unit; and providing a second virtual current source connected in parallel to the second VSM. The method further includes: using transient power P consumed by the first VSM. VSM1 A power-equivalent current reference is generated to control the second virtual current source, and the transient power P consumed by the second VSM is used. VSM2 A power-equivalent current reference is generated to control the first virtual current source.

[0070] Multiple implementation schemes may preferably implement the following features:

[0071] Preferably, the first terminal includes a first droop control unit, and the method includes connecting the output signal of the first droop control unit to the first virtual current source to simulate a first speed regulator.

[0072] Preferably, the second terminal includes a second droop control unit, and the method includes connecting the output signal of the second droop control unit to the second virtual current source to simulate a second speed regulator.

[0073] Preferably, the first terminal includes a first droop control unit, and the method includes connecting the output signal of the first droop control unit to the first PLL unit to simulate a first virtual mechanical speed regulator.

[0074] Preferably, the second terminal includes a second droop control unit, and the method includes connecting the output signal of the second droop control unit to the second PLL unit to simulate a second virtual mechanical speed regulator.

[0075] Preferably, the method includes applying a first factor K. VSM1 By using K VSM1 The output signal of the first virtual admittance is scaled proportionally to continuously change the characteristics of the first terminal from a voltage source to a current source.

[0076] Preferably, the method includes applying a second factor K. VSM2 By using K VSM2 The output signal of the second virtual admittance is scaled proportionally to continuously change the characteristics of the second terminal from a voltage source to a current source.

[0077] Preferably, the method includes taking the factor K VSM1 Feeding is made to at least one of the first droop control unit, the first virtual admittance unit, or the first PLL unit to increase or decrease the inertial response of the first terminal to interference in the first power grid.

[0078] Preferably, the method includes taking the factor K VSM2 Feeding is made to at least one of the second droop control unit, the second virtual admittance unit, or the second PLL unit to increase or decrease the inertial response of the second terminal to interference in the second power grid.

[0079] Preferably, the method includes using K VSM1 The output signal of the first virtual admittance is scaled proportionally using K. VSM1 The scaling gain of the first PLL unit is inversely scaled and / or K is used. VSM1 Scaling the first droop constant proportionally to the factor K VSM1 Scale the nominal power of the first VSM.

[0080] Preferably, the method includes using K VSM2 The output signal of the second virtual admittance is scaled proportionally using K. VSM2 The proportional gain of the second PLL unit is scaled inversely and / or using K VSM2 Scaling the second droop constant proportionally to the factor K VSM2 Scale the nominal power of the second VSM.

[0081] Preferably, the method includes using the factor K VSM1 To scale the rotation wheel factor, transient power, or inertia, or at least one of these.

[0082] Preferably, the method includes using the factor K VSM2 To scale the rotation wheel factor, transient power, or inertia, or at least one of these.

[0083] Preferably, the method includes providing a third terminal configured to be connected to a third power grid, the third terminal including a third current control unit, a third virtual admittance unit, and a third PLL unit; using the third virtual admittance unit and the third PLL unit to simulate the inertia of a third virtual synchronous machine (VSM); and providing a third virtual current source connected in parallel to the third VSM.

[0084] Preferably, the method includes using the transient power P consumed by the first VSM. VSM1 and the transient power P consumed by the second VSM VSM2 To generate a power-equivalent current reference to control the third virtual current source.

[0085] Preferably, the method includes using transient power PVSM1 consumed by the first VSM and transient power PVSM3 consumed by the third VSM to generate a power equivalent current reference to control the second virtual current source.

[0086] Preferably, the method includes using transient power PVSM2 consumed by the second VSM and transient power PVSM3 consumed by the third VSM to generate a power equivalent current reference to control the first virtual current source.

[0087] The methods and systems proposed in this disclosure have the following beneficial effects, among others:

[0088] - It can provide the characteristics of a virtual synchronous generator on both ends of the link without requiring excessive energy storage in the DC link.

[0089] - The amount of power grid support on each terminal can be easily adapted using a single factor.

[0090] - It can provide a smooth and seamless transition from voltage source characteristics to current source characteristics without changing the control mode.

[0091] - It can provide a smooth transition into and out of operation with maximum current without losing synchronization.

[0092] - Improved fault handling capabilities.

[0093] - Improved performance in weak and strong grids: Because active power setpoint tracking and regulator control are conceptually implemented with a fast current control loop, the performance degradation often observed in strong grids when using VSM control does not exist.

[0094] - The grid-forming vector current control used supports grid-connected operation as well as stand-alone grid-forming / islanding operation, without any changes to the control mode or parameters.

[0095] - Explicit current control and current limiting.

[0096] - Provides an explicit PLL to synchronize with the grid before closing the MCB and to prevent loss of synchronization during faults.

[0097] -PLL, together with virtual admittance, is used as a virtual oscillation equation, thus avoiding the use of power balancing for synchronization.

[0098] -Simplicity and reduced commissioning effort: Compared to conventional implementations with virtual synchronous generator control on a single terminal, using grid-formed vector current control can significantly reduce the amount of adjustment parameters required.

[0099] Several exemplary embodiments of this disclosure are described in detail below with reference to the accompanying drawings. The drawings are provided for illustrative purposes only and depict only exemplary embodiments of this disclosure to aid the reader's understanding of it.

[0100] Therefore, the accompanying drawings should not be considered as a limitation on the breadth, scope, or applicability of this disclosure.

[0101] It should be noted that these figures are not necessarily drawn to scale for clarity and ease of illustration.

[0102] Figure 1 An example of a schematic diagram of a grid interconnection / HVDC link according to existing technology is shown.

[0103] Figure 2 An example of a schematic diagram of a grid interconnection / HVDC link according to existing technology is shown.

[0104] Figure 3 An example of a schematic diagram of a grid interconnection / HVDC link according to existing technology is shown.

[0105] Figure 4 An example of a schematic diagram of a grid interconnection / HVDC link according to existing technology is shown.

[0106] Figure 5a and Figure 5b A schematic diagram according to this disclosure is shown.

[0107] Figure 6a and Figure 6b A schematic block diagram according to this disclosure is shown.

[0108] Figure 7The basic structure of power grid forming vector current control according to this disclosure is shown.

[0109] Figure 8 a to Figure 8 c illustrates a virtual machine variant scheme according to this disclosure.

[0110] Figure 9 a and Figure 9 b shows an equivalent model of the control scheme proposed according to this disclosure.

[0111] Figure 10 A vector diagram depicting the transient situation according to this disclosure is provided.

[0112] Figure 11 This illustrates the use factor k according to this disclosure. VSM To adjust the block diagram of the amount of power grid support.

[0113] Figure 12 A schematic diagram of a power grid forming vector current control system for controlling power grid interconnection according to the present disclosure is shown.

[0114] Figure 13 A schematic block diagram of a control system for terminals used for power grid interconnection according to the present disclosure is shown.

[0115] Figure 14 A schematic block diagram of a control system for terminals used for power grid interconnection according to the present disclosure is shown.

[0116] Figure 15 A schematic diagram of a power grid interconnection according to this disclosure is shown.

[0117] Figure 16 It shows the control based on Figure 15 A schematic diagram of a vector current control system formed by interconnected power grids.

[0118] Figure 17 A schematic illustration of the behavior of a grid-forming vector current control system used to control grid interconnection is shown in transient grid support mode.

[0119] Figure 18 A schematic illustration of the behavior of DC link voltage in transient grid support mode is shown.

[0120] Figure 19 A schematic illustration of the behavior of a grid-forming vector current control system used to control grid interconnection is shown in full grid support mode.

[0121] Figure 20 A schematic illustration of the behavior of DC link voltage in full grid support mode is shown.

[0122] Figure 5 schematically depicts a power grid forming vector current control (GFVCC) system according to this disclosure. Specifically, Figure 5a A virtual synchronous machine (VSM) with current sources connected in parallel is shown, while Figure 5b Its equivalent circuit diagram is shown. The system simulates the combined effects of a virtual machine (whose nominal power and grid-supported quantities can be scaled independently and online with a single input / parameter) and parallel-connected current sources. Therefore, the system can be controlled remotely. For certain aspects, such as fault handling, it is preferable to change the nominal power and grid-supported quantities within the application code.

[0123] More specifically, a disclosed GFVCC system according to an exemplary embodiment uses the control block shown in Figure 6 to simulate the proposed behavior. Its computation will be discussed below. Figure 6a Including the voltage V at the point of common coupling (PCC) PCC A power-fed phase-locked loop (PLL) block and a reference frequency f as input. ref Speed ​​control block. PLL output and V PCC The voltage V of the virtual synchronizer VSM Together, they are provided to the virtual admittance block. The speed control output, virtual admittance output, and reference current I are also included. ref Then, at the output voltage V out It is processed in the current control block.

[0124] Figure 6b Additionally, it includes a parameter k as an input to at least one of the PLL block, speed control block, and current control block. VSM Therefore, the amount of power grid support from the main power grid can be controlled. VSM It can also be used to smoothly change the characteristics of a control scheme from a voltage source to a current source, and also supports a combination of these two characteristics. A smooth transition to strict current control can be beneficial during grid faults to maintain synchronization with the grid. The control mode does not necessarily need to be changed.

[0125] The GFVCC system may also include an active voltage regulator (AVR). The GFVCC system according to this exemplary embodiment simulates the combined effects of virtual inertia, damper windings, virtual stator impedance, AVR, and droop regulator. A standard PLL block is used to simulate the virtual inertia and damper windings.

[0126] In contrast to conventional virtual machine implementations, active power setpoint tracking and droop regulators are conceptually implemented using fast current sources. Therefore, the performance degradation often observed when using VSM control is absent in strong grids. Consequently, good performance is achieved in both weak and strong grids. This method supports both unintended islanding and being islanded.

[0127] The complexity of this control scheme and the associated debugging workload are very low. By removing the cascaded voltage control loop and reusing the PLL to simulate inertia and damping, the number of parameters can be reduced from more than 15 in conventional methods to just 7 parameters with clear meaning and less interaction. Therefore, the system can also be tuned much more easily.

[0128] Another advantage of the proposed concept is that it simplifies the possibility of providing grid support features in applications with low energy storage—such as UPS in data centers, PVs with low or no storage, EV charging with buffered batteries, insensitive loads, etc.—by configuring the amount of grid support with a single parameter, and improving fault handling capabilities by smoothly transitioning into and out of operation with maximum current without losing synchronization.

[0129] The GFVCC system provides explicit current control and current limiting, as well as an explicit PLL for synchronizing with the grid before closing the MCB and for preventing synchronization loss during faults. The PLL, together with the virtual admittance, is used as a virtual swing equation, thus avoiding the need for synchronization using power balance.

[0130] Figures 7 to 11 A more detailed overview of the proposed power grid support control scheme and the corresponding equivalent circuit is shown.

[0131] exist Figure 5a and Figure 5b The concept describes Figure 7 The control scheme. Figure 5a The control scheme can be divided into two functional parts: a VSM (i.e., a rotating wheel or inertia, as shown by the dashed line) without any regulator, and a current source connected in parallel at the terminals of the virtual synchronous machine. A PLL is used to simulate the behavior of the virtual synchronous machine. The current source operates in parallel. It serves as both a fast power setpoint tracker and a regulator, as it ensures steady-state power balance between the grid and the rotating wheel. The inertial response of active power is ensured by virtual admittance. In steady state, all power flowing to the grid is injected by the current source. The rotating wheel only injects power during transient periods.

[0132] Figure 7A GVCC system according to one embodiment is shown, which has a droop control unit 1, a current control unit 2, a virtual admittance unit 3, and a phase-locked loop (PLL) unit 4. The virtual admittance unit 3 and the PLL unit 4 are configured to simulate the inertia of a VSM, and a virtual current source 5 is connected in parallel to the VSM. In this embodiment, the output signal of the droop control unit 1 is connected to the virtual current source 5 to simulate a speed regulator.

[0133] The GVCC system according to a preferred embodiment will now be described in more detail. The simplest form of the conventional advanced implementation of the virtual synchronizer is based on the oscillation equation. The goal of virtual synchronizer control is to make the behavior of the converter at the point of common coupling (PCC) equal to the behavior of the synchronizer.

[0134] Figure 8 a to Figure 8 c represents the equivalent circuit of the corresponding component, i.e., an advanced virtual machine variant scheme. Specifically, Figure 8 a shows the converter, Figure 8 b shows a VSM without virtual inductance, and Figure 8 c shows a VSM with a virtual inductance.

[0135] For simplicity, the following derivation is based on a converter with an inductive output filter. However, the control method proposed in this paper is applicable to converters with any type of output filter structure—such as L, LC, or LCL.

[0136] The swing equation will change the voltage source V C The frequency change is coupled with the active power flowing out of or into the machine. In the absence of a virtual inductance in the VSM, the converter voltage V... C This is equal to the back electromotive force (back-emf) of the virtual machine. Converter filter inductor L C It equals the stator inductance of the virtual machine, and the PCC voltage V PCC This is equal to the stator voltage of the virtual machine. Converter voltage V C The (virtual back electromotive force) has a rotational frequency ω. r To highlight its equivalence to the virtual rotor frequency. The PCC voltage (virtual stator voltage) has a rotational frequency ω. pcc The equation for the oscillation is as follows:

[0137]

[0138]

[0139] In quasi-steady state, the power flow is generated by the converter filter reactance X. C =ω N *L CVirtual back electromotive force V at both ends C and PCC voltage V PCC The angle difference between them is given.

[0140]

[0141] The portion spanning the transmission line exhibits a direct dependence on power flow, reactance, and angle difference, i.e.

[0142]

[0143] The oscillation equation therefore becomes

[0144]

[0145]

[0146] In the second step, according to Figure 8 c. Advanced VSM implementations have introduced virtual reactance X vs =ω N *L vs This virtual reactance slightly alters the equation. The converter voltage V in the phasor notation method... C Calculate using the following formula:

[0147] V C =V v -jX vs ·I c (6)

[0148] The virtual back electromotive force of the machine is designated as V. V (with rotation frequency ω) r and angle θ r And has moved to virtual reactance X vs After that. Regarding the behavior at PCC, the virtual machine now has X. v =X c +X vs The total virtual stator reactance, and the power flow equation changes to

[0149]

[0150] And the oscillation equation becomes

[0151]

[0152]

[0153] It should be noted that the virtual inductance L vs With converter filter inductor L C This increases the total virtual stator inductance of the virtual machine as seen from the PCC, thus adding to the series connection.

[0154] Figure 9 a and Figure 9 b shows an equivalent model of a control scheme according to an exemplary implementation. In this scheme, the oscillation equation is not directly implemented as in conventional methods. Furthermore, the effects of the PLL with virtual admittance are compared with the oscillation equation. Angle θ r Instead of being calculated using the oscillation equation, it is based on the PLL output of the PLL equation:

[0155]

[0156]

[0157] The derivative of equation (9) is derived as follows:

[0158]

[0159]

[0160] For small angle differences (which is the case of reasonable PLL tuning), the sine function can be approximated as sin(θ). r -θ pcc )≈(θ r -θ pcc ), and its derivative becomes sin(θ) r -θ pcc )≈ω r -ω pcc .

[0161] The system of equations can therefore be expressed by the following formula:

[0162]

[0163]

[0164] By comparing equation (11) with the original oscillation equation (1), the formal equivalent PLL gain of the two equations can be

[0165]

[0166] In a later step, a parallel current source will be used instead of the regulator power. Therefore, the proportional gain of the PLL simulates the damper winding effect, and the integral gain couples the angle difference with the increase / decrease in frequency, thus simulating the self-synchronization principle of a synchronous machine.

[0167] In contrast to the oscillating equation, the PLL does not implicitly couple the power flow to the angle difference, as it only describes the internal PLL angle θ. rThe evolution of this process. An additional current reference for the current controller must be used to explicitly ensure the coupling of power flow with the angle difference. This reference is created using a virtual admittance equivalent to the reciprocal of the virtual stator impedance, i.e.

[0168] Y v (s)=Z v (s) -1 (13)

[0169] The stator impedance is assumed to be of type RL. If this admittance is applied in the dq reference coordinate system, it can be described as follows:

[0170]

[0171]

[0172] Based on virtual admittance and the PLL reference coordinate system (V v =V N The virtual back electromotive force voltage V aligned with the d-direction of +j*0) V With PCC voltage V PCC (This is shown in the vector diagram depicting the transient situation) matrix multiplication of the voltage difference between the two points to create the current reference i. v，dq .

[0173] i v，d =Y v，dd (s)·(V pcc，d -V N )+Y v，dq (s)·V pcc，q (16)

[0174] i v，q =Y v，，qd (s)·(V pcc，d -V N )+Y v，qq (s)·V pcc，q (17)

[0175] Alternatively, it can also be applied to complex numbers and admittance can be applied according to the following formula.

[0176]

[0177] Under steady state and assuming R v <<ω N *L v Admittance simplified to

[0178]

[0179] This results in the following steady-state current reference:

[0180]

[0181]

[0182] If the PCC voltage V PCC Equal to nominal voltage V N Therefore, the injected q-axis current can be ignored. Thus, in the virtual stator reactance (X... v =ω N *L V The coupling between the voltage q and the active current in the d direction can be described as follows:

[0183]

[0184] Among them, the relevant power flow (assuming fast current tracking) is

[0185]

[0186] Therefore, according to this disclosure, the equivalence between the quasi-steady-state power flow regulated by the oscillation equation and the power flow created by the combination of PLL and virtual admittance is ensured.

[0187] Virtual admittance can be implemented using a fully dynamic model to ensure passivity and damping at high-frequency resonances. However, simplified admittance models (ignoring the s*L term) are also possible, as are more complex models that allow for additional filtering. Even asymmetric admittance is possible. The minimum requirement is to implement coupling between the q-voltage and the active current.

[0188] It has been shown that the combination of specific settings for PLL gain and virtual admittance allows the simulation of the effects of inertia, damper windings, and stator impedance of a virtual synchronous motor using conventional control structures for vector current control.

[0189] Below, the drooping block is further described. Unlike the original VSM implementation, the proposed control method only implements a rotating wheel without a regulator for mechanical input power. Therefore, the power injected from the rotating wheel in steady state (i v,d =0) will be zero, and the angle θ r With θ PCC Alignment.

[0190] The regulator power is replaced by a current source that ensures steady-state power balance. In addition to the reference from the virtual admittance, an active current reference is created according to the following formula.

[0191]

[0192] This results in the injected power being

[0193] P = Vpcc，d ·i d，ref =V pcc cos(θ r -θ pcc )·i d，ref =P set +K g ·(ω r -ω set (25)

[0194] The main difference is that this current reference is fed directly to the current controller without filtering the swing equation. Setpoint changes are fed through directly and are followed more quickly compared to the VSM implementation.

[0195] Given reasonable load conditions and reasonable virtual stator reactance for the rotating wheel, the angle difference θ r -θ pcc It is small, and cos(θ) r -θ pcc (Approaching one.) In this case, divide by V pcc,d It can be divided by the constant V N replace.

[0196] In addition, an active voltage regulator (AVR) can be added to maintain the magnitude of the PCC voltage constant. The AVR block can be of PI or integrator (I) type, or simply proportional (P) type. It creates a current reference i to regulate the voltage amplitude. qref .

[0197] i qref =G avr (s)-(|Y pcc |-V N (26)

[0198] However, since the virtual admittance has already implemented the proportional gain AVR effect using the following formula, the AVR block can also be omitted.

[0199]

[0200] As indicated above, the factor k can be used. VSM To adjust or scale the amount supported by the power grid. Preferably, the factor k VSM It is continuous and ranges from 0 to 1. Using this factor, the nominal power of a virtual machine that can be considered to operate in parallel with a current source having a single factor can be scaled. VSM The connections and inputs are shown in Figure 6b The equation is as follows:

[0201] Y virt =k VSM ·Yvirt，N (28)

[0202] M = k VSM ·M N (29)

[0203] K g =k VSM ·K g，N (30)

[0204]

[0205] refer to Figure 7 Modified block diagrams of the droop block, virtual admittance block, and PLL block are shown in Figure 11 middle.

[0206] Importantly, this makes the k used in the PLL block... VSM The reciprocal of the gain saturates to its maximum value to keep the PLL loop stable. Too high a proportional gain in the PLL loop can lead to instability—a well-known effect of standard vector current control with PLLs.

[0207] It should be noted that adjusting the grid-supported quantity by scaling the nominal power of the VSM is a unique feature of the GVCC control structure. The nominal power of the parallel current sources is unaffected, and the control structure can therefore continue to operate with normal current control and maximum power capacity, even if k VSM It was set to zero.

[0208] This feature can be achieved without conventional VSM implementations. With a standard VSM, power delivery capability is related to the VSM's nominal power. Power flow is controlled by advancing the angle of the virtual back electromotive force in the VSM and applying a voltage difference across the virtual stator reactance. Reducing the VSM's nominal power leads to an increase in this reactance, and therefore, an increased angle difference is required to deliver the same amount of power. Because the maximum angle difference is limited to 90°, the power delivery capability decreases with the VSM's nominal power. Therefore, in conventional control methods, the VSM's nominal power cannot be adjusted independently. Other applications besides those disclosed herein are also possible.

[0209] Figure 12 A schematic block diagram of a grid-forming vector current control system for controlling grid interconnection is shown. According to this disclosure, two terminals—a first terminal 100 connected to a first power grid A and a second terminal 200 connected to a second power grid B—are controlled using the aforementioned GFVCC. As discussed in detail above, this control scheme simulates a virtual generator, namely a virtual synchronous machine (VSM) 150 / 250 connected in parallel with (virtual) current sources 160 / 260.

[0210] The VSM section simulates the inertial response and low output impedance of a synchronous generator. However, it does not contribute to steady-state power exchange because it lacks the direct mechanical regulator simulation implemented in traditional virtual generator control schemes. To compensate for the missing regulator, controllable current sources 160 / 260 are actually connected in parallel to the virtual generator section / VSM 150 / 250.

[0211] As mentioned above, in this disclosure, both terminals 100 and 200 of the grid interconnection use GFVCC. The transient power consumed by the VSM section 250 controlled on the second terminal 200 is used as a feedforward term for the controllable current source 160 in the first terminal 100, i.e., as a power equivalent current reference. In the same manner, the transient power consumed by the virtual generator / VSM section 150 controlled on the first terminal 100 is used as a feedforward term for the controllable current source 260 on the second terminal 200, i.e., as a power equivalent current reference.

[0212] The setpoint is given as:

[0213]

[0214]

[0215]

[0216]

[0217]

[0218]

[0219]

[0220]

[0221]

[0222]

[0223] in:

[0224] V pcc1,d = The d-direction component of the PCC voltage measured at the first terminal of the power grid interconnection

[0225] V pcc1,q = The q-direction component of the PCC voltage measured at the first terminal of the power grid interconnection

[0226] V pcc2,d = The d-direction component of the PCC voltage measured at the second terminal of the power grid interconnection

[0227] V pcc2,q := The q-direction component of the PCC voltage measured at the second terminal of the power grid interconnection

[0228] i 1d,ref := The d-direction component of the current reference for the virtual current source portion used for control on the first terminal of the power grid interconnection

[0229] i 1q,ref := The q-direction component of the current reference for the virtual current source portion used for control on the first terminal of the power grid interconnection

[0230] i 2d,ref := The d-direction component of the current reference for the virtual current source portion used for control on the second terminal of the power grid interconnection

[0231] i 2q,ref := The q-direction component of the current reference for the virtual current source portion used for control on the second terminal of the power grid interconnection

[0232] P tr := Power transfer setpoint for steady-state power transferred between the first and second terminals of the power grid interconnection

[0233] P VSM1 = Power consumed by the VSM section controlled by the first terminal

[0234] P VSM2 = Power consumed by the VSM section controlled by the second terminal

[0235] = Current reference created using DC link voltage control

[0236] Q set,1 := Setpoint for injecting reactive power at the first terminal of the grid interconnection

[0237] Q set,2 := Setpoint for injecting reactive power at the second terminal of the grid interconnection

[0238] =The total d-direction current reference given to the current controller at the first terminal

[0239] =The total q-direction current reference given to the current controller at the first terminal

[0240] =The total d-direction current reference given to the current controller at the second terminal

[0241] =The total q-direction current reference given to the current controller at the second terminal

[0242] i v1,d := The d-direction component of the current reference created by the VSM section controlled on the first terminal

[0243] i v1,q := The q-direction component of the current reference created by the VSM section controlled on the first terminal

[0244] i v2,d := The d-direction component of the current reference created by the VSM section controlled on the second terminal

[0245] i v2,q := The q-direction component of the current reference created by the VSM section controlled on the second terminal

[0246] For simplicity, the DC link voltage control can be calculated based on the first terminal 100 only. However, it can also be shared between the first terminal 100 and the terminal 200.

[0247] Using only i v1d and i v2d —That is, the active current created by the virtual rotating wheel of the VSM—is used to calculate the feedforward term P. VSM2 and P VSM1 In steady state, both feedforward terms become zero because i v1d and i v2d The power becomes zero. The first published concept simulates only the inertial response at two terminals and obtains corresponding power from the opposite terminals to maintain power balance. The steady-state characteristics are determined by the power transfer setpoint P. tr and using Q set1 Q set2 Defined reactive power and simulated virtual generator (i v1q i v2q The reactive power is given.

[0248] In the implementation scheme below, two methods are discussed that provide not only transient grid support (inertial response) but also steady-state grid support.

[0249] Power transfer setpoint P tr It can be a quantity calculated by a slow top-level controller (secondary or tertiary control, power trading), or it can additionally contain a portion calculated using a droop curve supported by a steady-state grid based on the grid frequency at one terminal. Consider the grid frequencies at both terminals to determine P. tr More complex functions are also possible, as described below.

[0250] The current reference (I) of the virtual generator section of the control method v1,dq I v2,dq— This refers to the feedforward term, which is calculated using a PLL and virtual admittance to simulate the unique characteristics of the GFVCC in the virtual generator response. In the GFVCC, the nominal power of the virtual generator can be adjusted, and thus the response can be adjusted using a single factor K. VSM The amount supported by the power grid. It can be set individually for each terminal, a feature—K. VSM This feature can be used to reduce the amount of grid support on the strong grid side of the terminal so as not to interfere too much with the weak grid side. GFVCC has many implementation variations and options, such as current control in other reference coordinate systems, using K... VSM Improved fault handling, active damper branch in parallel with current source, and improved handling of asymmetric conditions.

[0251] Figure 13 A block diagram of a control system for the first terminal 100 is shown as an example. Those skilled in the art will understand that a similar control system is used for the second terminal. To this extent, according to... Figure 13 The control system disclosed in the implementation plan is similar to that in Figure 7 It relates to the block diagram of the control system depicted in the document.

[0252] Figure 13 A block diagram of a control system for a first terminal 100 according to one embodiment is shown. This control system includes a first current control unit 110, a first virtual admittance unit 120, and a phase-locked loop (PLL) unit 130. The first virtual admittance unit 120 and the PLL unit 130 are configured to simulate the inertia of the VSM, and a virtual current source 160 is connected in parallel to the VSM. In this embodiment, the transient power P consumed by the second VSM 250 of the second terminal 200... VSM2 The controller (not shown) of this control system is used to generate a power-equivalent current reference to control the first virtual current source. Furthermore, the transient power P consumed by the first VSM 150... VSM1 The controller of this control system uses a current reference to generate a power equivalent to control the second virtual current source 260 of the second terminal 200. Those skilled in the art will understand that a block diagram of the control system for the second terminal 100 according to one embodiment will appear corresponding. Furthermore, those skilled in the art will also understand that two... Figure 13 The control system of the embodiment described herein can be combined, one for the first terminal 100 and the second for the second terminal 200, to form a control system for controlling power grid interconnection according to this disclosure.

[0253] According to one implementation scheme, a method for determining P is described below. trThe method. If the controller of a terminal (e.g., the second terminal 200) is equipped with a typical droop curve, that is, the terminal includes a droop control unit, then P tr By the drooping curve and its offset P set The steady-state power flow is determined.

[0254] The equation above becomes

[0255]

[0256]

[0257]

[0258]

[0259]

[0260]

[0261]

[0262]

[0263]

[0264]

[0265] Where P tr yes

[0266]

[0267] And among them:

[0268] P droop,2 := Steady-state power on the second terminal

[0269] K g := droop constant

[0270] ω g2 := The power grid frequency measured at the second terminal

[0271] ω set,2 := Frequency setting point for the second terminal

[0272] Feedforward term P VSM2 The steady-state power P at the second terminal 200, as determined by the droop curve, is required. droop,2 To extend this, the following features can be achieved using the method according to this implementation scheme:

[0273] - Terminal 100: Transient power grid support (inertial response)

[0274] - Second terminal 200: Full grid support, i.e., transient (inertial response) and steady-state.

[0275] Those skilled in the art will understand that the behavior on the terminals can also be interchanged.

[0276] According to another implementation scheme, the following describes a method for determining P. tr The method. If the controllers on both terminals are equipped with typical droop curves, then P tr It is determined by the difference between the steady-state power flow determined by the droop curve and the offset power.

[0277] The equation becomes

[0278]

[0279]

[0280]

[0281]

[0282]

[0283]

[0284]

[0285]

[0286]

[0287]

[0288]

[0289]

[0290] And among them:

[0291] P droop,1 =Steady-state power on the first terminal

[0292] ω g1 := The power grid frequency measured at the first terminal

[0293] ω set,1 := Frequency setpoint for the first terminal

[0294] Feedforward term P VSM2 and P VSM1 It needs to be extended using the power flow determined by the droop curve on the opposite terminals.

[0295] Steady-state transmission power P rt become

[0296] P tr =P droop，2 -P droop，1

[0297] Assume ω set,1 =ω set,2 This leads to the following steady-state power transfer:

[0298] P tr =P set，2 -P set，1 -K g ·(ω g2 -ω g1 )

[0299] By setting

[0300]

[0301] P can be obtained tr

[0302] P tr =P set，tr -K g ·(ω g2 -ω g1 )

[0303] exist Figure 14 The complete control scheme is described in the text.

[0304] Using the method according to this implementation scheme, the following features can be achieved on the two terminals:

[0305] Terminal 1: Full grid support, i.e., transient (inertial response) and steady-state.

[0306] - Terminal 2: Full grid support, i.e. transient (inertial response) and steady state.

[0307] Figure 14 A block diagram of a control system for a first terminal 100 according to one embodiment is shown. This control system includes a first current control unit 110, a first virtual admittance unit 120, and a phase-locked loop (PLL) unit 130. The first virtual admittance unit 120 and the PLL unit 130 are configured to simulate the inertia of the VSM, and a virtual current source 160 is connected in parallel to the VSM. In this embodiment, the power P consumed by the second VSM 250 of the second terminal 200... VSM2 The controller (not shown) of this control system is used to generate a power-equivalent current reference to control the first virtual current source. Furthermore, the power P consumed by the first VSM150...VSM1 The controller of this control system uses a power-equivalent current reference to control the second virtual current source 260 of the second terminal 200. Furthermore, according to this embodiment, the control system also includes a first droop control unit 140, and the output signal of the first droop control unit 140 is connected to the first virtual current source 160 to emulate the first speed regulator. In other words, the control system according to this embodiment can be used to determine P as described in one of the foregoing embodiments. tr It should be further understood that the control system for the second terminal 200 according to one embodiment will appear corresponding. Furthermore, those skilled in the art will understand that the two... Figure 14 The control system of the embodiment described herein can be combined, one for the first terminal 100 and a second for the second terminal 200, to form a control system for controlling power grid interconnection according to this disclosure. Those skilled in the art will also understand that, according to... Figure 13 A control system of the implementation scheme described herein can be coupled with, according to Figure 14 A combination of control systems of the implementation scheme described herein, for example, according to Figure 13 The control system of the embodiment described herein is used for the first terminal 100 and according to Figure 14 The control system of the embodiment described herein is used for the second terminal 200, and vice versa, to form a control system for controlling power grid interconnection according to this disclosure.

[0308] Figure 16 It shows a method for controlling having, for example Figure 15 The diagram shown illustrates a schematic block diagram of a vector current control system formed by interconnecting three terminals of a power grid. However, those skilled in the art will understand that this disclosure is not limited to three terminals, but may include any number of terminals. According to this disclosure, all three terminals—the first terminal 100 connected to the first power grid A, the second terminal 200 connected to the second power grid B, and the third terminal 300 connected to the third power grid C—are controlled using the aforementioned GFVCC. Therefore, the controller of the control system uses the transient power P consumed by the first VSM. VSM1 and the transient power P consumed by the second VSM VSM2 To generate a power-equivalent current reference to control the third virtual current source, using the transient power P consumed by the first VSM. VSM1 and the transient power P consumed by the third VSM VSM3 To generate a power-equivalent current reference to control the second virtual current source, and to use the transient power P consumed by the second VSM. VSM2 and the transient power P consumed by the third VSM VSM3 A power-equivalent current reference is generated to control the first virtual current source.

[0309] With three terminals, the system of equations is given as follows:

[0310]

[0311]

[0312]

[0313]

[0314]

[0315]

[0316]

[0317]

[0318]

[0319]

[0320]

[0321]

[0322]

[0323]

[0324]

[0325]

[0326]

[0327]

[0328] in

[0329] V pcc3,d = The d-direction component of the PCC voltage measured at the third terminal of the power grid interconnection.

[0330] V pcc3,q = The q-direction component of the PCC voltage measured at the third terminal of the power grid interconnection.

[0331] i 3d,ref := The d-direction component of the current reference for the virtual current source portion used for control on the third terminal of the power grid interconnection

[0332] PVSM3 = Power consumed by the VSM section controlled by the third terminal

[0333] i 3q,ref := The q-direction component of the current reference for the virtual current source portion used for control on the third terminal of the power grid interconnection

[0334] Q set,3 := Setpoint for injecting reactive power at the third terminal of the power grid interconnection

[0335] =The total d-direction current reference given to the current controller at the third terminal

[0336] =The total q-direction current reference given to the current controller at the third terminal

[0337] i v3,d := The d-direction component of the current reference created using the VSM section controlled on the third terminal

[0338] i v3,q := The q-direction component of the current reference created using the VSM section controlled on the third terminal

[0339] P droop,3 := Steady-state power on the third terminal

[0340] ω g3 := The power grid frequency measured at the third terminal

[0341] k 12 := Percentage of power consumed by the VSM section controlled by the first terminal, supplied from the second terminal

[0342] k 13 := Percentage of power consumed by the VSM section controlled by the first terminal, supplied from the third terminal

[0343] k 21 := Percentage of power consumed by the VSM section controlled by the second terminal, supplied from the first terminal

[0344] k 23 := Percentage of power consumed by the VSM section controlled by the second terminal, supplied from the third terminal

[0345] k 31 := Percentage of power consumed by the VSM section controlled by the third terminal, supplied from the first terminal

[0346] k 32 := Percentage of power consumed by the VSM section controlled by the third terminal, supplied from the second terminal

[0347] Importantly, the total power allocation ratio is 1:

[0348] k 12 +k 13 =1

[0349] k 21 +k 23 =1

[0350] k 31 +k 32 =1

[0351] Figure 17 A schematic illustration of the behavior of a grid-forming vector current control system for controlling a grid interconnect in transient grid support mode is shown. In this embodiment, the power delivery setpoint for the interconnect is set to zero. A first terminal 100 is exemplarily connected to a relatively strong grid A with an SCR of 40 (short-circuit ratio), and a second terminal 200 is exemplarily connected to a second grid B with an SCR of 20. At time t = 1 s, a load step is applied in grid A. The interconnect delivers transient power to the load step and receives power from grid B. At time t = 3 s, a load step with a similar absolute magnitude is applied in grid B. The interconnect delivers transient power to the load step and receives power from grid A. In both cases, the power flow decays to zero after a period of time because only transient power is exchanged. The interconnect delivers an inertial response at both terminals. A transparent link for the transient power flow has been established.

[0352] Figure 18 A schematic illustration of the DC link voltage behavior under transient grid support mode is shown. The two transients have only a minor impact on the DC link voltage, and the controller is able to maintain a stable DC link. Although a scaled-down converter is used at low voltage levels according to this embodiment, the same behavior can be expected for medium and high voltage settings.

[0353] Figure 19A schematic illustration of the behavior of a grid-forming vector current control system for controlling a grid interconnect in full grid support mode is shown. According to this embodiment, bilateral full grid support (inertial response and steady-state support) is provided. The power delivery setpoint for the interconnect is set to zero. A first terminal 100 is exemplarily connected to a relatively strong grid A with an SCR of 40, and a second terminal 200 is exemplarily connected to a grid B with an SCR of 20. At time t = 1 s, a load step is applied in grid A. The interconnect delivers transient and steady-state power to the load step and obtains power from grid B. At time t = 3 s, a load step with a similar absolute magnitude is applied in grid B. Because grid B is weaker than grid A, the load step in each cell is higher, and the grid frequency in grid B will drop to a lower value. Therefore, the power flow in the interconnect will reverse and begin to support grid B, which has a heavier load compared to grid A. The interconnect delivers transient and steady-state power to grid B and obtains power from grid A.

[0354] Figure 20 A schematic illustration of the DC link voltage behavior in full grid support mode is shown. The two transients have only a minor impact on the DC link voltage, and the controller is able to maintain a stable DC link. Although this embodiment uses a downscaling converter at low voltage levels, the same behavior can be expected for medium and high voltage settings.

[0355] Other aspects, features, and advantages will become clear from the foregoing summary of the invention and the following description—including the drawings and claims.

[0356] While the invention has been shown and described in detail in the accompanying drawings and the foregoing description, such illustrations and descriptions should be considered illustrative or exemplary rather than restrictive. It should be understood that changes and modifications can be made by those skilled in the art within the scope of the appended claims. Specifically, the invention covers additional embodiments having any combination of features from the different embodiments described above and below.

[0357] Furthermore, in the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plurality. A single unit can perform the function of several features recited in the claims. Terms such as "substantially," "about," "approximately," etc., in conjunction with attributes or values, particularly define the attribute or value precisely. Any reference numerals in the claims should not be construed as limiting the scope.

[0358] Figure Labels

[0359] First terminal 100

[0360] First current control unit 110

[0361] First Virtual Admittance Unit 120

[0362] First PLL Unit 130

[0363] First droop control unit 140

[0364] First VSM 150

[0365] First virtual current source 160

[0366] Second terminal 200

[0367] Second current control unit 210

[0368] Second Virtual Admittance Unit 220

[0369] Second PLL unit 230

[0370] Second droop control unit 240

[0371] Second VSM 250

[0372] Second virtual current source 260

[0373] First Power Grid A

[0374] Second power grid B

[0375] Third power grid C

[0376] Citation List

[0377] [1] J. Rocabert, A. Luna, F. Blaabjerg and P. Rodríguez, “Control of power converters in AC microgrids”, IEEE Transactions on Power Electronics, Vol. 27, No. 11, pp. 4734-4749, November 2012;

[0378] [2] S.D'Arco and JA Suul, “Virtual Synchronous Machines: Classification of Implementation Methods and Equivalence Analysis of Microgrid Droop Controllers”, IEEE Grenoble Conference 2013, Grenoble, 2013, pp. 1-7;

[0379] [3] Y. Sun, X. Hou, J. Yang, H. Han, M. Su and JM Guerrero, “A new perspective on droop control in AC microgrids”, IEEE Transactions on Industrial Electronics, Vol. 64, No. 7, pp. 5741-5745, July 2017;

[0380] [4] M. Zhang, X. Yuan and J. Hu, “Inertia and master frequency specifications for PLL synchronization of VSC HVDC when attached to an islanded AC system”, IEEE Transactions on Power Systems, Vol. 33, No. 4, pp. 4179-4188, July 2018;

[0381] [5] H. Liu and J. Sun, “Voltage stability and control of offshore wind farms with AC collection and HVDC transmission”, IEEE Power Electronics Emerging & Selected Topics, Vol. 2, No. 4, pp. 1181-1189, December 2014;

[0382] [6] JRPérez, JASuul, S.D'Arco, A.Rodriguez-Cabero and M.Prodanovic, “Virtual Synchronous Machine Control of VSC HVDC for Power System Oscillation Damping”, IECON 2018 – 44th IEEE Industrial Conference, Washington, D.C., 2018, pp. 6026-6031;

[0383] [7] J. Zhu, CD Booth, G. P. Adam, A. J. Roscoe and C. GB right, “Inertial Simulation Control Strategies for VSC-HVDC Transmission Systems”, IEEE Transactions on Power Systems, Vol. 28, No. 2, pp. 1277-1287, May 2013.

Claims

1. A power forming vector current control system for controlling an interconnection of power grids, the system comprising: a first terminal (100) configured to be connected to a first power grid (A), the first terminal (100) comprising a first current control unit (110), a first virtual admittance unit (120) and a first phase locked loop, PLL, unit (130), wherein the first virtual admittance unit (120) and the first phase locked loop, PLL, unit (130) are configured to emulate an inertia of a first virtual synchronous machine, VSM, (150) and a first virtual current source (160) is connected in parallel to the first virtual synchronous machine, VSM, (150); a second terminal (200) configured to be connected to a second power grid (B), the second terminal (200) comprising a second current control unit (210), a second virtual admittance unit (220) and a second phase locked loop, PLL, unit (230), wherein the second virtual admittance unit (220) and the second phase locked loop, PLL, unit (230) are configured to emulate an inertia of a second virtual synchronous machine, VSM, (250) and a second virtual current source (260) is connected in parallel to the second virtual synchronous machine, VSM, (250); and a controller configured to generate a power equivalent current reference to control the second virtual current source (260) using transient power P VSM1 a controller configured to generate a power equivalent current reference to control the second virtual current source (260) using transient power P VSM2 a controller configured to generate a power equivalent current reference to control the second virtual current source (260) using transient power P 2. The system according to claim 1, wherein the first terminal (100) further comprises a first droop control unit (140) and an output signal of the first droop control unit (140) is connected to the first virtual current source (160) to emulate a first speed governor, and wherein the second terminal (200) further comprises a second droop control unit (240) and an output signal of the second droop control unit (240) is connected to the second virtual current source (260) to emulate a second speed governor.

3. The system according to claim 1, wherein the first terminal (100) further comprises a first droop control unit (140) and an output signal of the first droop control unit (140) is connected to the first phase locked loop, PLL, unit (130) to emulate a first virtual mechanical speed governor, and wherein the second terminal (200) further comprises a second droop control unit (240) and an output signal of the second droop control unit (240) is connected to the second phase locked loop, PLL, unit (230) to emulate a second virtual mechanical speed governor.

4. The system according to claim 2 or 3, wherein a first factor K VSM1 configured to continuously change the characteristic of the first terminal (100) from a voltage source to a current source by scaling the output signal of the first virtual admittance with said first factor K VSM1 proportionally to the output signal of the first virtual admittance; and a second factor K VSM2 configured to continuously change the characteristic of the second terminal (200) from a voltage source to a current source by scaling the output signal of the second virtual admittance with the second factor K VSM2 proportionally to the output signal of the second virtual admittance.

5. The system according to claim 4, wherein at least one of the first droop control unit (140), the first virtual admittance unit (120), or the first phase-locked loop, PLL, unit (130) is configured to receive the first factor k VSM1 to increase or decrease an inertial response of the first terminal (100) to disturbances in the first power grid; and at least one of the second droop control unit (240), the second virtual admittance unit (220), or the second phase-locked loop, PLL, unit (230) is configured to receive the second factor k VSM2 to increase or decrease an inertial response of the second terminal (200) to disturbances in the second power grid; Where the first factor K is used VSM1 The output signal of the first virtual admittance is scaled proportionally using the first factor K. VSM1 The scaling gain of the first phase-locked loop (PLL) unit (130) is inversely scaled and the first factor K is used to scale the gain inversely. VSM1 Scale the first droop constant proportionally, using the first factor K. VSM1 To scale the nominal power of the first virtual synchronizer VSM (150); and Where the second factor K is used VSM2 The output signal of the second virtual admittance is scaled proportionally using the second factor K. VSM2 The scaling gain of the second phase-locked loop (PLL) unit (230) is inversely scaled and the scaling gain is increased by the second factor K. VSM2 Scale the second droop constant proportionally, using the second factor K. VSM2 To scale the nominal power of the second virtual synchronizer VSM (250).

6. The system of claim 4, wherein the first factor k VSM1 and the second factor k VSM2 scale at least one of a rotation wheel factor or a transient power or the inertia.

7. The system according to any one of claims 1 to 3, 5 to 6, further comprising: a third terminal configured to be connected to a third power grid, the third terminal comprising a third droop control unit, a third current control unit, a third virtual admittance unit and a third phase locked loop, PLL, unit, wherein the third virtual admittance unit and the third phase locked loop, PLL, unit are configured to emulate an inertia of a third virtual synchronous machine, VSM, and a third virtual current source is connected in parallel to the third virtual synchronous machine, VSM; and wherein the controller is configured to: using the transient power P consumed by the first virtual synchronous machine VSM (150) VSM1 and the transient power P consumed by the second virtual synchronous machine VSM (250) VSM2 generating a power equivalent current reference to control the third virtual current source, using the transient power P consumed by the first virtual synchronous machine VSM (150) VSM1 and the transient power P consumed by the third virtual synchronous machine VSM VSM3 generating a power equivalent current reference to control a second virtual current source (260), and using the transient power P consumed by the second virtual synchronous machine VSM (250) VSM2 and the transient power P consumed by the third virtual synchronous machine VSM VSM3 generating a power equivalent current reference to control the first virtual current source (160).

8. A grid forming vector current control method for controlling an interconnection of power grids, the method comprising: providing a first terminal (100) configured to be connected to a first power grid (A), the first terminal (100) comprising a first current control unit (110), a first virtual admittance unit (120) and a first phase locked loop, PLL, unit; emulating an inertia of a first virtual synchronous machine, VSM, with the first virtual admittance unit (120) and the first phase locked loop, PLL, unit (130); providing a first virtual current source (160) connected in parallel to the first virtual synchronous machine, VSM, (150); providing a second terminal (200) configured to be connected to a second power grid (B), the second terminal (200) comprising a second current control unit (210), a second virtual admittance unit (220) and a second phase locked loop, PLL, unit (230); emulating an inertia of a second virtual synchronous machine, VSM, with the second virtual admittance unit (220) and the second phase locked loop, PLL, unit (230); providing a second virtual current source (260) connected in parallel to the second virtual synchronous machine, VSM, (250); and using the transient power P consumed by the first virtual synchronous machine VSM (150) VSM1 generating a power equivalent current reference to control the second virtual current source (260), and using the transient power P consumed by the second virtual synchronous machine VSM (250) VSM2 generating a power equivalent current reference to control the first virtual current source (160).

9. The method according to claim 8, wherein the first terminal (100) further comprises a first droop control unit (140), and wherein the method further comprises connecting an output signal of the first droop control unit (140) to the first virtual current source (160) to emulate a first speed governor; and wherein the second terminal (200) further comprises a second droop control unit (240), and wherein the method further comprises connecting an output signal of the second droop control unit (240) to the second virtual current source (260) to emulate a second speed governor.

10. The method according to claim 8, wherein the first terminal (100) further comprises a first droop control unit (140), and wherein the method further comprises connecting an output signal of the first droop control unit (140) to the first phase locked loop, PLL, unit (130) to emulate a first virtual mechanical speed governor; and wherein the second terminal (200) further comprises a second droop control unit (240), and wherein the method further comprises connecting an output signal of the second droop control unit (240) to the second phase locked loop, PLL, unit (230) to emulate a second virtual mechanical speed governor.

11. The method according to claim 9 or 10, comprising: Applying the first factor K VSM1 To achieve this by using the first factor K VSM1 The output signal of the first virtual admittance is scaled proportionally to continuously change the characteristics of the first terminal (100) from a voltage source to a current source; and Applying the second factor K VSM2 To achieve this by using the second factor K VSM2 The output signal of the second virtual admittance is scaled proportionally to continuously change the characteristics of the second terminal (200) from the voltage source to the current source.

12. The method according to claim 11, comprising: The first factor K VSM1 Feeding to at least one of the first droop control unit (140), the first virtual admittance unit (120), or the first phase-locked loop (PLL) unit (130) to increase or decrease the inertial response of the first terminal (100) to interference in the first power grid; The second factor K VSM2 Feeding to at least one of the second droop control unit (240), the second virtual admittance unit (220), or the second phase-locked loop (PLL) unit (230) to increase or decrease the inertial response of the second terminal (200) to disturbances in the second power grid; By using the first factor K VSM1 The output signal of the first virtual admittance is scaled proportionally using the first factor K. VSM1 The scaling gain of the first phase-locked loop (PLL) unit (130) is inversely scaled, and the first factor K is used to scale the gain of the first PLL unit (130) inversely. VSM1 Scale the first droop constant proportionally, using the first factor K. VSM1 Scaling the nominal power of the first virtual synchronizer VSM (150); and scaling the output signal of the second virtual admittance proportionally with the second factor K VSM2 scaling the output signal of the second virtual admittance proportionally with the second factor K VSM2 scaling the proportional gain of the second phase-locked loop, PLL, unit (230) inversely proportionally, and with the second factor K VSM2 scaling the second droop constant proportionally with the second factor K VSM2 scaling the nominal power of the second virtual synchronous machine, VSM, (250).

13. The method according to claim 11, comprising: scaling at least one of a rotation wheel factor or transient power or inertia with the first factor K VSM1 scaling at least one of a rotation wheel factor or transient power or inertia with the second factor K VSM2 scaling at least one of a rotation wheel factor or transient power or inertia with the first factor K 14. The method according to any one of claims 8 to 10, 12 to 13, further comprising: providing a third terminal configured to be connected to a third power grid, the third terminal comprising a third current control unit, a third virtual admittance unit and a third phase locked loop, PLL, unit; emulating an inertia of a third virtual synchronous machine, VSM, with the third virtual admittance unit and the third phase locked loop, PLL, unit; a third virtual current source connected in parallel to the third virtual synchronous machine VSM is provided; using the transient power P consumed by the first virtual synchronous machine VSM (150) VSM1 and the transient power P consumed by the second virtual synchronous machine VSM (250) VSM2 generating a power equivalent current reference to control the third virtual current source, using the transient power P consumed by the first virtual synchronous machine VSM (150) VSM1 and the transient power P consumed by the third virtual synchronous machine VSM VSM3 generating a power equivalent current reference to control the second virtual current source (260), and using the transient power P consumed by the second virtual synchronous machine VSM (250) VSM2 and the transient power P consumed by the third virtual synchronous machine VSM VSM3 generating a power equivalent current reference to control the first virtual current source (160).

Images