A grid-connected converter cluster model predictive dual-mode control method and system

Abstract

The application belongs to the technical field of renewable energy grid connection, and provides a grid-connected converter cluster model predictive dual-mode control method and system. By introducing a distributed compensation mechanism in the traditional control architecture, the influence of line impedance difference is overcome, the accurate reactive power sharing among converters in the grid-forming mode is realized, and the power oscillation hidden danger is eliminated. The finite control set model predictive control is used to replace the traditional PI controller as the unified inner loop, which effectively improves the smoothness and safety of the mode switching process. Through the integration of cooperative switching, power coordination and fast inner loop control, a complete cluster control architecture is constructed, and the operation stability of the system under wide range of grid impedance changes is comprehensively improved.

Description

Technical Field

[0001] This invention belongs to the field of renewable energy grid connection technology, specifically relating to a model prediction dual-mode control method and system for grid-connected converter clusters. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] As the proportion of renewable energy generation, such as wind and solar power, in the power system continues to increase, their operational stability is crucial to grid security. Currently, the mainstream control strategies for grid-connected converters include grid-following (GFL) control and grid-forming (GFM) control. With the increasing penetration rate of intermittent renewable energy sources such as wind and solar power, the equivalent impedance of the grid exhibits wide-range fluctuations, and the system short-circuit ratio decreases significantly, creating a weak or even extremely weak grid operating environment. This problem intensifies the dynamic interaction between grid-connected converters and the grid, sharply increasing the risk of system instability and posing a severe challenge to maintaining the stable and efficient operation of converter clusters. Therefore, it is urgent to study advanced converter control methods that can adapt to large fluctuations in grid impedance to ensure the safe and stable operation of grid-connected systems with a high proportion of renewable energy.

[0004] Currently, the industry's strategies for improving converter stability mainly follow two technical paths. The first path focuses on optimization within a single control mode. For example, for grid-based control, it widens the stability boundary under small-signal models by reshaping the phase-locked loop or introducing active damping; or for grid-based control, it designs transient energy dissipation methods under large disturbances based on Lyapunov stability theory.

[0005] The second approach is a grid impedance adaptive dual-mode switching strategy that combines grid tracking and grid construction. The core of this approach lies in real-time online identification of the grid impedance or short-circuit ratio (e.g., by injecting disturbance signals or utilizing the system's inherent harmonics and employing algorithms such as recursive least squares for estimation), and then dynamically switching the operating mode based on the identification results: when the system short-circuit ratio is high and the grid is determined to be strong, the converter operates in grid tracking mode to pursue optimal power point tracking accuracy; when the short-circuit ratio is lower than a set threshold and the grid is determined to be weak, it switches to grid construction mode to provide voltage and frequency support for the system. This strategy achieves a preliminary match between the operating mode and grid conditions.

[0006] However, when dealing with multi-converter parallel cluster systems, existing solutions are entirely based on single-machine control. Their design only considers the interaction between a single converter and the grid, lacking coordinated planning for "grid switching" at the multi-machine system level. When multiple independently operating dual-mode control converters are connected in parallel, their judgments of the grid state and mode decisions are decentralized and asynchronous. This lack of coordinated autonomous behavior is itself a potential cause of system instability. Furthermore, existing solutions lack effective cluster power coordination strategies. Especially after switching to grid-connected mode, traditional droop control struggles to achieve precise reactive power distribution due to differences in line parameters, affecting the system's steady-state performance and voltage quality. The traditional PI controllers commonly used in their inner loops are limited by bandwidth, resulting in insufficient response speed and anti-interference capabilities during the dynamic process of mode switching. Summary of the Invention

[0007] To address the aforementioned problems, this invention proposes a model predictive dual-mode control method and system for grid-connected converter clusters. By using finite control set model predictive control as a unified inner loop and combining it with a cluster reactive power sharing mechanism, this invention effectively solves the challenges of coordinated control and power allocation during dual-mode switching in multi-machine systems, significantly improving the dynamic performance and operational stability of the system.

[0008] According to some embodiments, the present invention adopts the following technical solution: A model predictive dual-mode control method for grid-connected converter clusters includes the following steps: Perform converter current inner loop control: Establish a discrete model of the converter for prediction, transform the predicted current to the dq rotating coordinate system through Park transformation, traverse all possible switching states of the converter, calculate the corresponding predicted current for each state, calculate the cost function based on the predicted current, select the switching state that minimizes the cost function value, and apply it to the next control cycle. Perform grid connection / grid connection mode switching control: acquire voltage and current signals of the point of common coupling, determine grid strength, and when the grid strength is lower than a preset threshold, switch at least a portion of the converters to grid connection mode; when the grid strength rises to a higher threshold, switch at least a portion of the converters to grid connection mode. After the mode switching, synchronize the internal state of the converters with the target mode. Reactive power compensation and voltage recovery control in network mode: Compare the reactive power output of each converter with the average value of the cluster to generate a compensation signal to eliminate distribution errors; generate a voltage recovery signal based on the deviation of the voltage of each node from the average value of the cluster to maintain the system voltage within the allowable range.

[0009] As an alternative implementation method, the converter current inner loop control process is performed, which is applicable to both grid-following control and grid-connected control modes.

[0010] As an alternative implementation, the process of establishing a discrete model of the converter for prediction includes: in αβ In a stationary coordinate system, based on the inductor-voltage equation, the predicted current at the next sampling time is calculated as follows: ; i 1 αβ ( k+ 1) These are the predicted next time intervals ( k+ The converter output measured at time 1 α and β Components in coordinate system i 1 αβ ( k ) are the measurement times ( k The converter output measured at (time) α and β Components in coordinate system R f This is the equivalent series resistance of the filter inductor. L f This is the value of the filter inductance. T s The sampling period of the control system, u αβ ( k The measurements are: (i) the measurement time, (ii) the measurement time, and (iii) the measurement time. k The converter-side output voltage measured at constant time is α and β Components in coordinate system u c αβ ( k The measurements are: (i) the measurement time, (ii) the measurement time, and (iii) the measurement time. k The grid-side voltage or point of common coupling voltage measured at any time α and β Components in coordinate system.

[0011] As an alternative implementation, the process of calculating the cost function based on the predicted current includes: the cost function is: ,in, This is the reference value for the d-axis current on the grid side. This is the predicted value of the d-axis current on the grid side. This is the reference value for the q-axis current on the grid side. This is the predicted value of the q-axis current on the grid side.

[0012] As an alternative implementation, the process of obtaining the voltage and current signals at the point of common coupling (PCC) and determining the grid strength includes obtaining the PCC voltage and current signals, calculating the system's equivalent short-circuit ratio, and using this as a criterion for grid strength. The calculation formula is as follows: ; in, S sc For system short-circuit capacity, P rated This refers to the rated power of the converter cluster.

[0013] As an alternative implementation, after mode switching, the process of synchronizing the converter's internal state with the target mode includes phase reference pre-synchronization, voltage reference pre-synchronization, and current reference pre-synchronization. When the switching signal changes from grid-following type to grid-building type, the system's inner-loop reference signal changes from the output signal of the grid-following controller to the output signal of the grid-building controller after pre-synchronization. When the switching signal changes from grid-building type to grid-following type, the system's inner-loop reference signal changes from the output signal of the grid-building controller to the output signal of the grid-following controller after pre-synchronization, thus achieving shock-free mode switching.

[0014] As a further limitation, the phase reference pre-synchronization is achieved by adjusting the angular frequency of the virtual synchronizer, and the pre-synchronization phase is obtained by integrating the angular velocity: ; ω GFM The original angular frequency of the virtual synchronous generator. θ grid The actual phase of the grid voltage. θ GFM The original phase of the virtual synchronous generator. ω GFM_pre The angular frequency of the virtual synchronous generator after pre-synchronization adjustment. θ GFM_pre This refers to the phase of the virtual synchronous generator after pre-synchronization.

[0015] As a further defined implementation, the voltage reference pre-synchronization stage is as follows: ; u oGFM This is the original output voltage reference value for the virtual synchronous generator. u ogrid This is the actual measured value of the power grid voltage. u oGFMpre This is the reference value for the output voltage after pre-synchronization; The current reference pre-synchronization element is as follows: ; i 2 dq The dq-axis component of the grid-side current. u o dqFor the dq-axis components of the grid voltage, u oGFMpre dq The dq-axis components of the pre-synchronization voltage. ω o The rated angular frequency, i 1GFMpre dq The dq-axis component of the inner loop current reference quantity after pre-synchronization.

[0016] As an alternative implementation method, the process of reactive power compensation and voltage recovery control in grid-connected mode includes: the output voltage reference value of each converter is composed of the traditional droop characteristic and the compensation signal. ; U oi For the first i The output voltage of the converter U o i As the output voltage reference of the converter, n i For the first i The reactive power droop factor of the converter. Q i For the output of each converter, δ u It is a composite compensation signal;

[0017] δ qc This is a reactive power sharing signal. δ ur For voltage recovery signal, Q k For the output of each converter, E k The output voltage of each converter. n This represents the number of converters.

[0018] A grid-connected converter cluster model predictive dual-mode control system includes: The converter current inner loop control module is configured to establish a discrete model of the converter for prediction, transform the predicted current to the dq rotating coordinate system through Park transformation, traverse all possible switching states of the converter, calculate the corresponding predicted current for each state, calculate the cost function based on the predicted current, select the switching state that minimizes the cost function value, and apply it to the next control cycle. The grid connection / grid connection mode switching control module is configured to acquire the voltage and current signals of the common connection point, determine the grid strength, and switch at least a portion of the converters to grid connection mode when the grid strength is lower than a preset threshold; when the grid strength rises to a higher threshold, at least a portion of the converters are switched to grid connection mode. After the mode switching is performed, the internal state of the converter is synchronized with the target mode. In the network configuration mode, the reactive power compensation and voltage recovery control module is configured to compare the reactive power output of each converter with the average value of the cluster, generate a compensation signal to eliminate distribution errors, and generate a voltage recovery signal based on the deviation between the voltage of each node and the average value of the cluster to maintain the system voltage within the allowable range.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention establishes a cluster collaborative decision-making mechanism to ensure that all converters perform synchronous mode switching based on a unified grid state assessment, fundamentally avoiding the risk of system instability caused by decentralized decision-making. Simultaneously, a cluster power coordination strategy deeply coupled with dual-mode operation is designed. By introducing a distributed compensation mechanism into the traditional control architecture, the influence of line impedance differences is overcome, achieving precise reactive power sharing among converters in the grid configuration mode, thus eliminating potential power oscillations. Furthermore, a finite control set model predictive control is used to replace the traditional PI controller as the unified inner loop, leveraging its rapid dynamic response and precise constraint handling capabilities to effectively improve the smoothness and safety of the mode switching process. Finally, by integrating collaborative switching, power coordination, and fast inner-loop control, a complete cluster control architecture is constructed, comprehensively improving the system's operational stability under a wide range of grid impedance variations.

[0020] This invention enables cooperative mode switching of converter clusters, improving system stability; solves the problem of difficult reactive power distribution in network mode, achieving precise reactive power allocation; adopts model predictive control as a unified inner loop, improving the dynamic performance of the system; and constructs a complete cluster control architecture, enhancing the system's adaptability and robustness.

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0022] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0023] Figure 1 This is a control block diagram for a traditional grid-connected converter. Figure 2 This is a control block diagram for a traditional grid-type converter. Figure 3 This is a topology diagram of a grid-connected converter cluster circuit according to one embodiment; Figure 4 This is a flowchart illustrating the principle of finite set model predictive control in one embodiment. Figure 5 This is a block diagram of the pre-synchronization and mode switching signal logic of one embodiment; Figure 6 This is an improved block diagram of reactive power sharing and voltage recovery control according to one embodiment. Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0026] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0027] Where there is no conflict, the embodiments and features described in this application may be combined with each other.

[0028] Example 1 As described in the background section, with the increasing penetration of renewable energy in power systems, power generation clusters consisting of multiple grid-connected converters are becoming increasingly common. Under weak grid conditions with large fluctuations in grid impedance, the stable operation of such clusters faces severe challenges: firstly, traditional grid-following control modes based on phase-locked loops are prone to instability under weak grid conditions; secondly, it is difficult to achieve precise reactive power sharing among multiple converters, leading to circulating currents and system oscillations; and thirdly, existing control methods lack adaptive mode switching capabilities when grid conditions change, resulting in slow dynamic response.

[0029] To address the aforementioned issues, this embodiment proposes a grid-connected converter cluster model predictive dual-mode control method oriented towards active grid support. By using finite control set model predictive control as a unified inner loop and combining it with a cluster reactive power sharing mechanism, this method effectively solves the challenges of coordinated control and power allocation during dual-mode switching in multi-machine systems, significantly improving the dynamic performance and operational stability of the system.

[0030] The following is a detailed introduction.

[0031] Converter Models and Control Equations in Grid-following and Grid-connecting Modes Figure 1 and 2 These are control models for grid-connected and grid-connected converters, respectively. The grid-connected converter is connected to the power grid via an LC filter. Based on the power calculation formula in dq coordinates, we can obtain the instantaneous power injected by the converter to the point of common coupling (PCC): (1) In grid-connected control mode, the controller synchronizes with the grid via a phase-locked loop and acts as a controlled current source. Its dq-axis current reference values ​​are directly calculated from active and reactive power commands. (2) The above formula P and Q These are references for the active and reactive power outputs of the converter, respectively. i d and i q These are the reference values ​​for the inner current loop. In particular, the phase-locked loop can achieve precise synchronization in steady state, making... u c q =0, the above expression can be simplified to: (3) Then, through the current inner loop control based on model predictive control, the grid-connected converter can be controlled.

[0032] For grid-type control mode, phase-locked loops are not used to orient the grid voltage to achieve power angle tracking. Instead, virtual synchronous machine technology is used to simulate the rotor motion equation of a synchronous generator to establish the frequency and phase of the converter output voltage. The replication of the output voltage is regulated through reactive power-voltage droop control, as shown in the following expression: (4) in, J For virtual inertia, D The damping coefficient is... K p and K q These are the active and reactive power droop factors, respectively. The resulting voltage amplitude... E and phase θ This serves as a reference command for the voltage outer loop and the converter phase angle.

[0033] Compared to the grid control mode, the grid control mode adds a voltage outer loop, using the voltage signal generated by the droop control as the input reference command to generate the reference command for the current inner loop.

[0034] Model predictive control-based converter current inner loop control framework Finite control set model predictive control is adopted as a unified inner-loop current controller, applicable to both grid-following and grid-connected control modes. First, a discretized converter model is established for prediction. αβ In a stationary coordinate system, based on the inductor-voltage equation, the predicted current at the next sampling moment can be derived as follows: (5) i 1 αβ ( k+ 1) These are the predicted next time intervals ( k+ The converter output measured at time 1 α and β Components in coordinate system i 1 αβ ( k ) are the measurement times ( k The converter output measured at (time) α and β Components in coordinate system R f This is the equivalent series resistance of the filter inductor. L f This is the value of the filter inductance. T s The sampling period of the control system, u αβ ( k ) are the measurement times ( k The converter-side output voltage measured at (time) α and β Components in coordinate system u c αβ ( k ) are the measurement times ( k The grid-side voltage (or point of common coupling voltage) measured at (time) α and β Components in coordinate system.

[0035] By using the Park transformation, the predicted current is transformed to the dq rotating coordinate system, resulting in... i 1 dq ( k+1) Then, the controller checks all eight possible switching states of the converter, calculates the corresponding predicted current for each state, and evaluates it using the following cost function: (6) Ultimately, the cost function is chosen to be... g The switch state with the smallest value is directly applied to the next control cycle. This method eliminates the need for a traditional PWM modulator, achieving fast and accurate current tracking. Figure 4 It also demonstrates in detail all aspects of the implementation of the current inner loop predictive control.

[0036] Adaptive Tracking / Construction Mode Switching Strategy Based on Grid Impedance By assessing grid strength in real time and coordinating the synchronous switching of operating modes across the entire converter cluster, the system first calculates the equivalent short-circuit ratio (SCR) of the system by monitoring the voltage and current signals at the point of common coupling in real time. This SCR is then used as a criterion for grid strength. The calculation formula is as follows: (7) in, S sc For system short-circuit capacity, P rated This refers to the rated power of the converter cluster.

[0037] For a single-unit converter, when the calculated SCR is lower than a preset threshold, the cluster control unit generates a unified "switch to grid-connected mode" command; when the SCR rises to a higher threshold, it generates a "switch to grid-following mode" command. The logic for switching commands for multi-unit converters is similar (taking a three-converter unit as an example): (8) Upon receiving the switching command, the system initiates a pre-synchronization process. This process synchronizes the converter's internal state with the target mode through the following adjustments: phase reference pre-synchronization is achieved by adjusting the angular frequency of the virtual synchronizer, and the pre-synchronization phase is obtained by integrating the angular velocity. (9) ω GFM The original angular frequency of the virtual synchronous generator. θ grid The actual phase of the grid voltage (measured via PLL). θ GFM The original phase of the virtual synchronous generator. ω GFM_pre The angular frequency of the virtual synchronous generator after pre-synchronization adjustment. θ GFM_pre This refers to the phase of the virtual synchronous generator after pre-synchronization.

[0038] The voltage reference pre-synchronization process is as follows: (10) u oGFM This is the original output voltage reference value for the virtual synchronous generator. u ogrid This is the actual measured value of the power grid voltage. u oGFMpre This is the reference value for the output voltage after pre-synchronization.

[0039] The current reference pre-synchronization element is as follows: (11) i 2 dq The dq-axis component of the grid-side current. u o dq For the dq-axis components of the grid voltage, u oGFMpre dq The dq-axis components of the pre-synchronization voltage. ω o The rated angular frequency, i 1GFMpre dq The dq-axis component of the inner loop current reference quantity after pre-synchronization.

[0040] When the switching signal is switched from the tracking network type to the network-connected type, the switching signal controller will change the system inner loop reference signal from the output of the tracking network controller. i 1GFL , θ PLL Switching to pre-synchronized network controller output i 1GFMpre , θ GFMpre This achieves a shockless mode switching. The same mechanism is used in the reverse switching process to ensure that it is consistent with the grid status before the switch, thus achieving a truly seamless mode switching.

[0041] Power compensation mechanism considering reactive power compensation and voltage recovery To achieve accurate reactive power distribution and voltage stability of converter clusters in grid-connected mode, this invention employs a composite compensation strategy, where the output voltage reference value of each converter is composed of both traditional droop characteristics and compensation signals: (12) U oi For the first i The output voltage of the converter U oi As the output voltage reference of the converter, n i For the first i The reactive power droop factor of the converter. Q i For the output of each converter, δ u It is a composite compensation signal.

[0042] (13) δ qc This is a reactive power sharing signal. δ ur For voltage recovery signal, Q k For the output of each converter, E k The output voltage of each converter. n This represents the number of converters.

[0043] This control structure compares the reactive power output of each converter in real time. Q i Generate a compensation signal based on the cluster average. δ qc To eliminate allocation errors; and simultaneously monitor the voltage of each node. E i The deviation from the cluster average, through δ ur The system voltage is maintained within the allowable range. The synergistic effect of the two compensation terms ensures accurate reactive power distribution while avoiding voltage instability caused by overcompensation.

[0044] Example 2 A grid-connected converter cluster model predictive dual-mode control system includes: The converter current inner loop control module is configured to establish a discrete model of the converter for prediction, transform the predicted current to the dq rotating coordinate system through Park transformation, traverse all possible switching states of the converter, calculate the corresponding predicted current for each state, calculate the cost function based on the predicted current, select the switching state that minimizes the cost function value, and apply it to the next control cycle. The grid connection / grid connection mode switching control module is configured to acquire the voltage and current signals of the common connection point, determine the grid strength, and switch at least a portion of the converters to grid connection mode when the grid strength is lower than a preset threshold; when the grid strength rises to a higher threshold, at least a portion of the converters are switched to grid connection mode. After the mode switching is performed, the internal state of the converter is synchronized with the target mode. In the network configuration mode, the reactive power compensation and voltage recovery control module is configured to compare the reactive power output of each converter with the average value of the cluster, generate a compensation signal to eliminate distribution errors, and generate a voltage recovery signal based on the deviation between the voltage of each node and the average value of the cluster to maintain the system voltage within the allowable range.

[0045] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of one or more computer-usable storage media (including, but not limited to, disk storage, etc.) containing computer-usable program code. CD - ROM It takes the form of a computer program product implemented on (such as optical memory, etc.).

[0046] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0047] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0048] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0049] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art without creative effort within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A model predictive dual-mode control method for grid-connected converter clusters, characterized in that, Includes the following steps: Perform converter current inner loop control: Establish a discrete model of the converter for prediction, transform the predicted current to the dq rotating coordinate system through Park transformation, traverse all possible switching states of the converter, calculate the corresponding predicted current for each state, calculate the cost function based on the predicted current, select the switching state that minimizes the cost function value, and apply it to the next control cycle. Perform grid connection / grid connection mode switching control: acquire voltage and current signals of the point of common coupling, determine grid strength, and when the grid strength is lower than a preset threshold, switch at least a portion of the converters to grid connection mode; when the grid strength rises to a higher threshold, switch at least a portion of the converters to grid connection mode. After the mode switching, synchronize the internal state of the converters with the target mode. Reactive power compensation and voltage recovery control in grid-connected mode: Compare the reactive power output of each converter with the average value of the cluster, generate a compensation signal to eliminate distribution errors; Based on the deviation of each node's voltage from the cluster average, a voltage recovery signal is generated to maintain the system voltage within the allowable range; The process of reactive power compensation and voltage recovery control in grid-connected mode includes: the output voltage reference value of each converter is composed of the traditional droop characteristic and the compensation signal. ； U oi For the first i The output voltage of the converter U o i * As the output voltage reference of the converter, n i For the first i The reactive power droop factor of the converter. Q i For the output of each converter, δ u It is a composite compensation signal; δ qc This is a reactive power sharing signal. δ ur For voltage recovery signal, E k The output voltage of each converter. n This represents the number of converters.

2. The grid-connected converter cluster model predictive dual-mode control method as described in claim 1, characterized in that, The converter current inner loop control process is applicable to both grid-following control and grid-connected control modes.

3. The grid-connected converter cluster model predictive dual-mode control method as described in claim 1, characterized in that, The process of establishing a discrete model of the converter for prediction includes: αβ In a stationary coordinate system, based on the inductor-voltage equation, the predicted current at the next sampling time is calculated as follows: ； i 1 αβ ( k+ 1) These are the predicted next time intervals ( k+ The converter output measured at time 1 α and β Components in coordinate system i 1 αβ ( k ) are the measurement times ( k The converter output measured at (time) α and β Components in coordinate system R f This is the equivalent series resistance of the filter inductor. L f This is the value of the filter inductance. T s The sampling period of the control system, u αβ ( k The measurements are: (i) the measurement time, (ii) the measurement time, and (iii) the measurement time. k The converter-side output voltage measured at constant time is α and β Components in coordinate system u c αβ ( k The measurements are: (i) the measurement time, (ii) the measurement time, and (iii) the measurement time. k The grid-side voltage or point of common coupling voltage measured at any time α and β Components in coordinate system.

4. The grid-connected converter cluster model predictive dual-mode control method as described in claim 1, characterized in that, The process of calculating the cost function based on the predicted current includes: The cost function is: ,in, This is the reference value for the d-axis current on the grid side. This is the predicted value of the d-axis current on the grid side. This is the reference value for the q-axis current on the grid side. This is the predicted value of the q-axis current on the grid side.

5. The grid-connected converter cluster model predictive dual-mode control method as described in claim 1, characterized in that, The process of obtaining voltage and current signals at the point of common coupling (PCC) to determine grid strength includes obtaining the PCC voltage and current signals, calculating the system's equivalent short-circuit ratio (EMR), and using this as a criterion for grid strength. The calculation formula is as follows: ； in, S sc For system short-circuit capacity, P rated This refers to the rated power of the converter cluster.

6. The grid-connected converter cluster model predictive dual-mode control method as described in claim 1, characterized in that, After mode switching, the process of synchronizing the converter's internal state with the target mode includes phase reference pre-synchronization, voltage reference pre-synchronization, and current reference pre-synchronization. When the switching signal changes from grid-following type to grid-building type, the system's inner loop reference signal changes from the output signal of the grid-following controller to the output signal of the grid-building controller after pre-synchronization. When the switching signal changes from grid-building type to grid-following type, the system's inner loop reference signal changes from the output signal of the grid-building controller to the output signal of the grid-following controller after pre-synchronization, thus achieving shock-free mode switching.

7. The grid-connected converter cluster model predictive dual-mode control method as described in claim 6, characterized in that, The phase reference pre-synchronization is achieved by adjusting the angular frequency of the virtual synchronizer, and the pre-synchronization phase is obtained by integrating the angular velocity: ； ω GFM The original angular frequency of the virtual synchronous generator. θ grid The actual phase of the grid voltage. θ GFM The original phase of the virtual synchronous generator. ω GFM_pre The angular frequency of the virtual synchronous generator after pre-synchronization adjustment. θ GFM_pre This refers to the phase of the virtual synchronous generator after pre-synchronization.

8. The grid-connected converter cluster model predictive dual-mode control method as described in claim 6, characterized in that, The voltage reference pre-synchronization link is as follows: ； u oGFM This is the original output voltage reference value for the virtual synchronous generator. u ogrid This is the actual measured value of the power grid voltage. u oGFMpre This is the reference value for the output voltage after pre-synchronization; The current reference pre-synchronization element is as follows: ； i 2 dq The dq-axis component of the grid-side current. u o dq The dq-axis component of the grid voltage. u oGFMpre dq The dq-axis components of the pre-synchronization voltage. ω o The rated angular frequency, i 1GFMpre dq The dq-axis component of the inner loop current reference quantity after pre-synchronization.

9. A grid-connected converter cluster model predictive dual-mode control system, characterized in that, include: The converter current inner loop control module is configured to establish a discrete model of the converter for prediction, transform the predicted current to the dq rotating coordinate system through Park transformation, traverse all possible switching states of the converter, calculate the corresponding predicted current for each state, calculate the cost function based on the predicted current, select the switching state that minimizes the cost function value, and apply it to the next control cycle. The grid connection / grid connection mode switching control module is configured to acquire the voltage and current signals of the common connection point, determine the grid strength, and switch at least a portion of the converters to grid connection mode when the grid strength is lower than a preset threshold; when the grid strength rises to a higher threshold, at least a portion of the converters are switched to grid connection mode. After the mode switching is performed, the internal state of the converter is synchronized with the target mode. The reactive power compensation and voltage recovery control module in the grid mode is configured to compare the reactive power output of each converter with the average value of the cluster and generate a compensation signal to eliminate distribution errors. Based on the deviation of each node's voltage from the cluster average, a voltage recovery signal is generated to maintain the system voltage within the allowable range; The process of reactive power compensation and voltage recovery control in grid-connected mode includes: the output voltage reference value of each converter is composed of the traditional droop characteristic and the compensation signal. ； U oi For the first i The output voltage of the converter U o i * As the output voltage reference of the converter, n i For the first i The reactive power droop factor of the converter. Q i For the output of each converter, δ u It is a composite compensation signal; δ qc This is a reactive power sharing signal. δ ur For voltage recovery signal, E k The output voltage of each converter. n This represents the number of converters.

Images