Modeling method of multi-converter coupling system and electronic equipment

Abstract

The invention discloses a modeling method of a multi-converter coupling system and electronic equipment. The modeling method provided by the invention realizes multiple beneficial effects by adopting a complex vector node admittance matrix unified integration technology under an alpha-beta static coordinate system, firstly, the method avoids necessary global coordinate transformation and phase-locked loop dependence under a traditional dq coordinate system, the modeling process is remarkably simplified, and the calculation complexity is reduced; secondly, heterogeneous control characteristics of the following network type converter and the constructed network type converter are uniformly represented through a second-order complex vector structure, the broadband coupling effect among the voltage, the current and the frequency is directly revealed, and an accurate model basis is provided for analyzing the multi-mode oscillation problem of a hybrid system; and finally, the admittance matrix has structural characteristics and expandability, an efficient maintenance basis is provided for dynamic updating during subsequent topology change, and the practical value of the model in real-time stability analysis and control of the new energy power plant is greatly improved.

Description

Technical Field

[0001] This invention relates to the technical field of power plant impedance modeling, specifically to a modeling method and electronic equipment for a multi-converter coupled system. Background Technology

[0002] In related technologies, with the rapid development of new energy sources such as wind power and photovoltaics, new energy power plants have occupied an increasingly higher proportion of the power system. Traditional power systems centered on synchronous generators are gradually being replaced by high-proportion power electronic interface systems, leading to a decrease in system inertia, weakened voltage support capacity, and significant changes in operating characteristics. Under high penetration conditions, the dynamic behavior of new energy power plants has become a key factor determining the stability of the power system.

[0003] In related technologies, the modeling methods for hybrid systems of new energy power plants and grid-type and grid-type converters rely on global coordinate transformation and cannot uniformly represent the voltage-current-frequency broadband coupling effect. This leads to technical problems in adapting to topology changes in scenarios with dynamic topology changes. Summary of the Invention

[0004] The technical problem this invention aims to solve is that, in related technologies, the modeling methods for hybrid systems of new energy power plants and grid-type and grid-connected converters rely on global coordinate transformation and cannot uniformly represent the voltage-current-frequency broadband coupling effect. This leads to a technical problem of being unable to adapt to topology changes in scenarios with dynamic topology variations. The purpose is to provide a modeling method and electronic equipment for multi-converter coupled systems, solving the technical problem of being unable to adapt to topology changes.

[0005] This invention is achieved through the following technical solution:

[0006] In a first aspect, the present invention provides a modeling method for a multi-converter coupled system, comprising:

[0007] Obtain the electrical parameters of the passive network in the power plant, and the control parameters of the grid-connected converter and the grid-connected converter;

[0008] Based on the electrical and control parameters, a broadband coupled admittance network model of the power plant is constructed in a first coordinate system, wherein the first coordinate system is a stationary coordinate system used to characterize the instantaneous characteristics of AC electrical quantities; wherein the broadband coupled admittance network model integrates the output admittance models of the grid-connected converter and the grid-connected converter with the admittance model of the passive network through a unified second-order complex vector structure, so as to directly characterize the voltage-current-frequency broadband coupling effect caused by converter control interaction.

[0009] Furthermore, the unified second-order complex vector structure characterizes the frequency coupling effect in the following way:

[0010] The output admittance model of the grid-connected converter is configured such that its output current disturbance vector is equal to the product of its own admittance function and the terminal voltage disturbance vector, plus a coupling term; wherein the coupling term is composed of the product of the frequency-shifted admittance function, the rotation factor, and the conjugate of the terminal voltage disturbance vector.

[0011] The output admittance model of the grid converter is configured to be expressed in an isomorphic form consistent with the grid converter model, and its admittance function reflects the power droop and voltage-current dual-loop control characteristics of the grid converter.

[0012] Furthermore, the admittance model of the passive network is represented by a complex vector node admittance matrix, which is configured as follows:

[0013] For a network with n nodes, the complex vector node admittance matrix is ​​a 2n×2n block diagonal matrix. Its upper left n×n submatrix represents the admittance relationship of the node at the main frequency, and its lower right n×n submatrix represents the admittance relationship of the node at the coupling frequency after frequency shift.

[0014] Furthermore, integrating the output admittance models of the grid-connected converter and the network-type converter with the admittance model of the passive network includes:

[0015] Add the self-admittance element in the output admittance model of the grid converter to the main diagonal position of the main frequency block and the coupling frequency block corresponding to the node connected to the grid converter in the complex vector node admittance matrix of the passive network.

[0016] Add the self-admittance element in the output admittance model of the grid converter to the main diagonal position of the main frequency block and the coupling frequency block corresponding to the node connected by the grid converter in the complex vector node admittance matrix of the passive network;

[0017] In the output admittance models of the grid-type converter and the network-type converter, the cross admittance elements representing the frequency coupling effect are added to the corresponding non-main diagonal positions in the complex vector node admittance matrix, respectively. The cross admittance elements include admittance functions corrected by frequency offset and rotation factor. Through the addition operation, the wideband coupling admittance network model is formed, and the block diagonal structure of the complex vector node admittance matrix is ​​maintained.

[0018] Secondly, the present invention provides a method for updating a broadband coupled admittance network model, comprising:

[0019] Monitor topology change events in the power plant; wherein, the topology change events include the commissioning or disconnection of converters or the switching on or off of electrical connections;

[0020] In response to detected topology change events, a pre-stored broadband coupled admittance network model correction strategy is invoked;

[0021] Based on the correction strategy of the pre-stored broadband coupled admittance network model, the broadband coupled admittance network model of the power plant is updated to maintain its accurate representation of the dynamic characteristics of the power plant under the current operating conditions; wherein, the broadband coupled admittance network model is obtained by the modeling method of the multi-converter coupled system described above.

[0022] Further, the step of updating the broadband coupled admittance network model of the power plant according to the correction strategy based on the pre-stored broadband coupled admittance network model includes:

[0023] In response to the topology change event, a matrix correction operation corresponding to the event type is performed; wherein the matrix correction operation is configured as follows:

[0024] If the topology change event is the addition of a new converter or node, then in the complex vector node admittance matrix of the wideband coupled admittance network model, a row and column corresponding to the new converter or node are added, and the new admittance element is initialized based on its electrical parameters.

[0025] If the topology change event is the removal of a converter or node, then all rows and columns corresponding to the removed converter or node are deleted from the complex vector node admittance matrix.

[0026] If the topology change event is an electrical connection line disconnection, then the mutual admittance elements between the nodes connected to the disconnected line in the complex vector node admittance matrix are set to zero, and the self-admittance elements of the connected nodes are corrected simultaneously.

[0027] Thirdly, the present invention provides an impedance stability analysis method for a new energy power plant, comprising:

[0028] Establish a wideband coupled admittance network model or obtain an updated wideband coupled admittance network model; wherein, the wideband coupled admittance network model is constructed using the modeling method for a multi-converter coupled system described above; wherein, the updated wideband coupled admittance network model is obtained by updating the wideband coupled admittance network model using the update method described above.

[0029] The equivalent impedance characteristics of the power generation field are determined based on the broadband coupled admittance network model.

[0030] A stability criterion analysis was performed on the equivalent impedance characteristics to assess the stability risk of the power plant over a wide frequency range.

[0031] Furthermore, after establishing or obtaining the broadband coupled admittance network model, a verification step for the model is also included:

[0032] At a predetermined node in the power plant, a predetermined harmonic disturbance current signal is injected.

[0033] Based on the broadband coupled admittance network model, the harmonic voltage response of one or more nodes in the power plant is calculated;

[0034] The calculated harmonic voltage response is compared with the reference voltage response obtained through dynamic simulation or physical measurement.

[0035] The accuracy of the broadband coupled admittance network model is verified based on the comparison results. After the verification is passed, the stability criterion analysis of the equivalent impedance characteristics is performed to assess the stability risk of the power plant in a broadband range.

[0036] Fourthly, the present invention provides an electronic device, comprising: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, the instructions being executed by the one or more processors to cause the one or more processors to implement the method described above.

[0037] Fifthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0038] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0039] The modeling method provided by this invention achieves multiple beneficial effects by employing a unified integration technique of complex vector node admittance matrices in the αβ stationary coordinate system. First, this method avoids the necessary global coordinate transformation and phase-locked loop dependence in the traditional dq coordinate system, significantly simplifying the modeling process and reducing computational complexity. Second, by uniformly representing the heterogeneous control characteristics of grid-connected and grid-connected converters through a second-order complex vector structure, it directly reveals the broadband coupling effect between voltage, current, and frequency, providing an accurate model foundation for analyzing multimodal oscillation problems in hybrid systems. Finally, the admittance matrix possesses structured characteristics and scalability, providing an efficient maintenance basis for dynamic updates during subsequent topology changes (e.g., converter switching or line faults), greatly enhancing the practical value of the model in real-time stability analysis and control of new energy power plants. Attached Figure Description

[0040] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0041] In the attached diagram:

[0042] Figure 1 A flowchart illustrating a modeling method for a multi-converter coupled system provided in the embodiments of this specification;

[0043] Figure 2 A schematic diagram of the overall structure of a hybrid system of grid-connected and grid-connected converters for a new energy power plant, provided in an embodiment of the present invention;

[0044] Figure 3 A flowchart for verifying the voltage distribution results of grid-connected converters and grid-connected converters during mixed operation, as provided in an embodiment of the present invention;

[0045] Figure 4 The verification result diagram shows the voltage distribution results of the grid-connected converter and the grid-connected converter operating in combination, as provided in the embodiment of the present invention. Figure 4 'a' represents the verification of the voltage amplitude distribution at node 1. Figure 4 b represents the voltage phase distribution verification at node 1;

[0046] Figure 5 One of the schematic diagrams of the complex vector node admittance matrix for adding a new converter and adding, reducing or disconnecting nodes provided in the embodiments of the present invention;

[0047] Figure 6 This is the second schematic diagram of the complex vector node admittance matrix for adding a new converter and adding, reducing or disconnecting nodes, as provided in the embodiments of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0049] In related technologies, with the rapid development of new energy sources such as wind power and photovoltaics, the penetration rate of new energy power plants in the power system has significantly increased, gradually replacing traditional synchronous generators as the main power source. This transformation leads to a decrease in system inertia, a weakening of voltage support capability, and profound changes in operating characteristics.

[0050] New energy power plants are often connected to the power grid via power electronic converters (GFMs). These converters can be categorized into two types: grid-following (GFL) and grid-forming (GFM) converters. GFL converters track the grid voltage phase through phase-locked loops (PLLs) and rely on the external grid for frequency reference, exhibiting current source characteristics. GFM converters, on the other hand, actively establish voltage and frequency references through mechanisms such as voltage controllers and droop control, exhibiting voltage source characteristics. In practical engineering, both types of converters are often connected to the same power plant in a hybrid configuration, forming a complex coupled system with the external grid. This hybrid operation mode means that the system's dynamic response is affected by multiple control mechanisms. Especially in scenarios with weak grids or high proportions of new energy connections, the system may experience multimodal oscillations over a wide frequency range (e.g., low, medium, and high frequencies), seriously threatening the safe and stable operation of the power plant and the power system.

[0051] To analyze the stability of such systems, impedance modeling methods are often employed. These methods are typically based on a dq rotating coordinate system, treating a single converter as an equivalent impedance source and combining it with grid impedance for interactive analysis. For example, in the dq coordinate system, the impedance model of a grid-connected converter involves linearizing the relationship between small disturbance voltage and current variables, and handling frequency coupling effects through phase-locked loops and coordinate transformations. Similarly, the model of a grid-connected converter also relies on rotating coordinate transformations and the dynamic coupling of the power control loop.

[0052] However, these methods have inherent drawbacks: First, they heavily rely on global coordinate transformations (e.g., conversion from dq to αβ via phase-locked loops), leading to complex model calculations and introducing additional frequency coupling, making it difficult to intuitively reveal the system-level broadband interaction mechanism. Second, multi-machine systems are often simplified to single-machine equivalent models, failing to accurately characterize the heterogeneous control characteristics and interactions between different types of converters (e.g., GFL and GFM). Finally, while the sequence impedance method partially solves the frequency coupling problem, it still relies on local phase information and coordinate transformations, making it difficult to handle the dynamic evolution of hybrid systems within a unified framework. Especially when topology changes occur (e.g., converter connection / disconnection, line switching), the model lacks scalability and dynamic update capabilities, resulting in stability assessment results that are seriously inconsistent with reality.

[0053] The root of these technical problems lies in the fact that existing modeling methods cannot directly and uniformly characterize the broadband coupling effect between voltage, current, and frequency without avoiding coordinate transformations. Specifically, methods relying on the dq coordinate system introduce hidden frequency coupling terms (e.g., through conjugate operations and rotation factors), causing the model to lose frequency interaction information between different nodes. Simultaneously, the model structure is rigid and cannot flexibly adapt to dynamic changes in system topology, thus limiting its application in real-time stability analysis and control. Therefore, there is an urgent need for a new modeling method that can uniformly characterize the coupling dynamics of grid-type and mesh-type converters without relying on global coordinate transformations, and supports rapid model updates during topology changes.

[0054] This embodiment provides a modeling method for a multi-converter coupled system. The execution entity of the method can be a power plant monitoring computer, which can be deployed in the central control room of a new energy power plant (such as a wind farm or a photovoltaic power plant). It collects electrical parameters of the passive network (collector lines, transformers, filters, etc.) within the power plant, as well as the control system parameters of each converter (including GFLs and GFMs). Using this data, it runs the modeling method to construct a broadband coupled admittance network model of the entire power plant, which is used for real-time stability assessment, oscillation mode analysis, and control parameter optimization at the power plant level.

[0055] The method can also be implemented by a server, such as the analysis server of a power grid dispatch center. To assess the stability of the regional power grid after a high proportion of renewable energy is integrated, the dispatch center requires accurate power plant models. The dispatch center's server can obtain the necessary parameters from multiple power plants and execute this modeling method to establish a broadband stability analysis model of the regional power grid that includes detailed models of multiple power plants.

[0056] like Figure 1 As shown, the method may include:

[0057] Step S12: Obtain the electrical parameters of the passive network in the power plant, and the control parameters of the grid-connected converter and the grid-connected converter.

[0058] In this embodiment, the execution entity can receive, read, or collect the basic data necessary for constructing the broadband coupled admittance network model from internal memory or external system through its input / output interface or data processing unit.

[0059] Specifically, the electrical parameters of the passive network may include:

[0060] Line parameters, which are the distributed parameters of power cables or overhead lines connecting converter clusters, transformers, and combiner stations, may include: resistance, inductance, and capacitance to ground per unit length.

[0061] Transformer parameters are the equivalent circuit parameters of a voltage transformer or other transformers in the power plant. Specifically, they include winding resistance, leakage reactance, magnetizing reactance, and turns ratio.

[0062] Passive filter parameters refer to the component parameters of an LC or LCL filter installed to suppress specific harmonics. Specifically, these parameters include the inductance of the filter reactor, the capacitance of the capacitor, and their equivalent series resistance.

[0063] Topological connections are represented by the electrical connections between the passive components, specifying which components are connected in series or parallel, and the correspondence between each node (i.e., electrical connection point). For example, it can be a network connection list or topology diagram that defines all nodes and branches.

[0064] In this embodiment, the control parameters of the grid-connected converter may include:

[0065] Phase-locked loop (PLL) control parameters, which may include the proportional gain and integral gain of the PLL proportional-integral controller, determine the dynamic response speed and accuracy of the converter in synchronizing with the grid voltage.

[0066] Current loop control parameters may include the proportional and integral gains used in the inner loop current tracking controller, as well as parameters for any possible cross-coupling compensation terms.

[0067] Feedforward compensation parameters, which can be used to set the gain of the grid voltage feedforward term to improve dynamic response performance.

[0068] In this embodiment, the control parameters of the grid converter may include:

[0069] The power outer loop control parameters can represent the droop coefficients in realizing the active-frequency droop and reactive-voltage droop control laws, as well as the time constant or cutoff frequency of the outer loop power calculation filter.

[0070] Voltage loop control parameters, which can represent the proportional-integral gain of the outer loop voltage controller, which is used to generate the inner loop current reference value.

[0071] Virtual impedance parameters can be represented as virtual resistance and virtual inductance values ​​artificially introduced to enhance stability.

[0072] The inner loop current control parameters can represent the proportional-integral gain of the inner loop current controller.

[0073] In this embodiment, during the design or planning phase, the aforementioned parameters can be directly retrieved by the executing entity from a pre-stored power plant design database or from technical manuals provided by the equipment manufacturer. During the real-time operation phase, the executing entity can also receive uploaded data from field sensors or equipment controllers in real time through the communication interface with the power plant monitoring and data acquisition system.

[0074] Step S14: Based on the electrical parameters and control parameters, construct a broadband coupled admittance network model of the power plant in the first coordinate system, wherein the first coordinate system is a stationary coordinate system used to characterize the instantaneous characteristics of AC electrical quantities; wherein the broadband coupled admittance network model integrates the output admittance models of the grid-connected converter and the grid-connected converter with the admittance model of the passive network through a unified second-order complex vector structure, so as to directly characterize the voltage-current-frequency broadband coupling effect caused by the converter control interaction.

[0075] In this embodiment, the first coordinate system is a two-dimensional orthogonal coordinate system fixed in space, used to directly represent the instantaneous values ​​and phases of AC electrical quantities (voltage, current) without requiring rotational transformations dependent on the system frequency. Specifically, the first coordinate system can be an αβ stationary coordinate system. This coordinate system is connected to the natural three-phase abc coordinate system through a Clarke transformation, with its α-axis coinciding with the a-phase axis and its β-axis leading the α-axis by 90 electrical degrees.

[0076] In this embodiment, the wideband coupled admittance network model can be represented as a system-level model with a complex vector node admittance matrix as its core data structure. Its wideband aspect means that the model maintains accuracy across a wide frequency range from subsynchronous to high frequencies, and its coupled aspect means that it can characterize the interaction between voltage, current, and frequency. In a specific implementation, this model can be configured as a high-order complex matrix. For a network with n physical nodes, the dimension of this admittance matrix is ​​2n×2n. Its structure is organized as a block diagonal matrix, where the upper left n×n submatrix (main frequency block) directly characterizes the node admittance relationship at the network fundamental frequency and its nearby frequency bands (main frequency); while the lower right n×n submatrix (coupled frequency block) characterizes the admittance after frequency offset (e.g., offset from the main frequency by -2). The admittance relationship at the coupling frequency of ).

[0077] In this embodiment, the unified second-order complex vector structure can be represented as a unified mathematical form for describing the voltage-current disturbance relationship at the output port of a single converter (whether GFL or GFM). It can be a 2x2 complex matrix that encompasses the dynamic relationship of positive and negative order coupling within a compact framework.

[0078] In a specific implementation scheme, the output admittance model of any converter can be uniformly expressed in the following form:

[0079] The converter's output current disturbance vector equals a term consisting of the converter's own admittance function, plus a coupling term characterizing frequency coupling. This coupling term is derived from the frequency offset (…). The admittance function after () and a fixed rotation factor () The product of the frequency and the conjugate vector of the terminal voltage disturbance constitutes the model. This structure allows the model to explicitly reveal that for a frequency of , When a positive-sequence voltage disturbance occurs, the converter will not only respond to a frequency of The positive sequence current will also generate a frequency of ( - The negative sequence coupling current. The difference between grid-type and mesh-type converters lies only in the specific expressions of each admittance function element in the unified structure matrix, which are determined by the control parameters obtained in step S12.

[0080] In this embodiment, the executing entity can mathematically integrate multiple distributed converter output admittance models with a unified passive network admittance model into a single, complete system model. Specifically, the self-admittance element in each converter model can be added to the main diagonal position of the passive network matrix corresponding to the node connected to that converter (present simultaneously in the main frequency block and the coupling frequency block). The cross admittance element (i.e., the admittance function corresponding to the coupling term) in each converter model can be added to the non-main diagonal position of the passive network matrix related to the node connected to that converter.

[0081] In this embodiment, due to the adoption of the first coordinate system and the unified second-order complex vector structure, the off-diagonal block matrix elements in the broadband coupled admittance network model directly and clearly correspond to the strength of frequency coupling. Analysts can directly observe or calculate the model's response at different frequencies without performing complex coordinate inverse transformations or order decompositions. This allows for the direct interpretation of broadband stability issues caused by heterogeneous control interactions such as phase-locked loop dynamics and power droop control, for example, impedance comparison and Nyquist stability analysis at specific frequencies. This provides a precise theoretical basis for the stable operation of the system and the optimization of controller parameters.

[0082] The modeling method provided in this embodiment achieves multiple beneficial effects by employing a unified integration technique of complex vector node admittance matrices in the αβ stationary coordinate system: First, this method avoids the necessary global coordinate transformation and phase-locked loop dependence in the traditional dq coordinate system, significantly simplifying the modeling process and reducing computational complexity; Second, by uniformly representing the heterogeneous control characteristics of grid-connected and grid-connected converters through a second-order complex vector structure, it directly reveals the broadband coupling effect between voltage, current, and frequency, providing an accurate model foundation for analyzing multimodal oscillation problems in hybrid systems; Finally, the admittance matrix possesses structured characteristics and scalability, providing an efficient maintenance foundation for dynamic updates during subsequent topology changes (e.g., converter switching or line faults), greatly enhancing the practical value of the model in real-time stability analysis and control of new energy power plants.

[0083] In some implementations, the unified second-order complex vector structure characterizes the frequency coupling effect in the following way:

[0084] The output admittance model of the grid-connected converter is configured such that its output current disturbance vector is equal to the product of its own admittance function and the terminal voltage disturbance vector, plus a coupling term; wherein the coupling term is composed of the product of the frequency-shifted admittance function, the rotation factor, and the conjugate of the terminal voltage disturbance vector.

[0085] In this embodiment, the output current disturbance vector can be expressed as the linearized change in the output current caused by a small change in the AC side output voltage of the converter (i.e., the terminal voltage disturbance vector) in the first coordinate system. Specifically, it can be a column vector with two elements, which comprehensively characterizes the dynamic response of the converter output current in terms of amplitude and phase.

[0086] In this embodiment, the self-admittance function can be expressed as the dominant admittance characteristic of the converter output port in the frequency domain, reflecting the connection voltage disturbance and current response, without considering frequency coupling effects. Specifically, it can be an element in a transfer function matrix of complex frequency variables, and its specific form can be determined by the main circuit parameters of the converter, such as the current loop controller and filters. For example, for a grid-connected converter, it mainly reflects the dynamic characteristics of its current control loop.

[0087] In this embodiment, the coupling term can represent an additional term in the unified second-order complex vector structure specifically used to quantify frequency coupling effects.

[0088] Specifically, the admittance function after frequency shift can be expressed as a new function obtained by complex frequency shifting its own admittance function. Mathematically, this operation corresponds to the sideband effect caused by periodic modulation (e.g., phase-locked loop tracking, power oscillation) in a physical system. Specifically, this can be achieved by replacing the Laplace operator with (...) in the frequency domain expression of the admittance function. The function obtained after offsetting (or other offset forms).

[0089] The rotation factor, which can be represented as a fixed complex phase rotation operator, is used to achieve a phase transformation at a certain angle. Specifically, it can be... ,in, This is the initial phase angle or reference phase of the system voltage vector. This factor serves as a bridge connecting the positive-sequence and negative-sequence components.

[0090] The conjugate of the terminal voltage perturbation vector can be expressed as taking the complex conjugate of the terminal voltage perturbation vector. This mathematical operation physically relates the positive-sequence perturbation to the negative-sequence response. It can be a column vector that introduces frequency components opposite to the original perturbation sequence.

[0091] The output admittance model of the grid converter is configured to be expressed in an isomorphic form consistent with the grid converter model, and its admittance function reflects the power droop and voltage-current dual-loop control characteristics of the grid converter.

[0092] In this embodiment, the models of the two types of converters can share the exact same second-order complex vector matrix equation form, that is, their output current disturbance vectors are equal to the sum of their own terms and the coupling terms defined above.

[0093] In this embodiment, the admittance function, reflecting the power droop and voltage-current dual-loop control characteristics of the grid-type converter, can be expressed as follows: under isomorphic mathematical form, the self-admittance function and the frequency offset admittance function in the coupling terms of the grid-type converter model may differ in their specific functional expressions from those of the grid-type converter. Specifically, the analytical expression of this admittance function is derived by linearizing the corresponding control loop (including the power calculation link, active-frequency / reactive-voltage droop control link, outer-loop voltage controller, and inner-loop current controller). Therefore, although the mathematical structure is isomorphic, the admittance value ultimately calculated by substituting it into the model contains the voltage source behavior, inertial response, and power synchronization characteristics of the grid-type converter.

[0094] This implementation standardizes and integrates the modeling framework by configuring output admittance models with a unified second-order complex vector structure for both grid-connected and multi-mode converters. This unified structure not only accurately characterizes the frequency coupling characteristics of grid-connected converters dynamically induced by the phase-locked loop (PLL), but also accurately reflects the dynamic response of multi-mode converters based on power droop and dual-loop voltage-current control through the differences in the admittance function. Thus, within a unified mathematical framework, it intuitively reveals the complex voltage-current-frequency broadband coupling mechanism during the hybrid operation of heterogeneous converters, laying a direct and clear model foundation for system-level stability analysis and control design.

[0095] In some implementations, the admittance model of the passive network is represented by a complex vector node admittance matrix, which is configured as follows:

[0096] For a network with n nodes, the complex vector node admittance matrix is ​​a 2n×2n block diagonal matrix. Its upper left n×n submatrix represents the admittance relationship of the node at the main frequency, and its lower right n×n submatrix represents the admittance relationship of the node at the coupling frequency after frequency shift.

[0097] In this embodiment, each element of the matrix can be a complex number, representing the equivalent admittance between nodes. In its construction, the positions of non-zero elements in the matrix are first determined based on the network topology (i.e., which nodes are connected via passive components such as lines and transformers). Then, based on the electrical parameters obtained in step S12 (e.g., line resistance, inductance, and capacitance to ground), the admittance values ​​of these passive components in the frequency domain are calculated and filled into the corresponding positions in the matrix. Self-admittance elements (diagonal elements) are set to the sum of the admittances of all branches connected to that node, while mutual admittance elements (off-diagonal elements) are set to the negative values ​​of the admittances of the branches connecting to the corresponding nodes.

[0098] In this embodiment, the 2n×2n dimensional block diagonal matrix can be represented as a description of it in the first coordinate system. Each physical node in this coordinate system requires two scalars (α component and β component) to fully describe its electrical state. Therefore, for n physical nodes, the dimension of its state vector is 2n, and the corresponding system matrix dimension is 2n×2n.

[0099] In this embodiment, the n×n submatrix in the upper left corner can characterize the conventional admittance characteristics of the passive network at the system fundamental frequency and its vicinity (i.e., the dominant frequency). It describes how a voltage disturbance at one node causes a current response at other nodes at the same frequency within that frequency band.

[0100] The n×n submatrix in the lower right corner represents the admittance characteristics of a passive network in the frequency domain (i.e., the coupling frequency) after a specific frequency offset.

[0101] In some embodiments, integrating the output admittance models of the grid-connected converter and the network-connected converter with the admittance model of the passive network includes:

[0102] Add the self-admittance element from the output admittance model of the grid converter to the main diagonal position of the main frequency block and the coupling frequency block corresponding to the node connected to the grid converter in the complex vector node admittance matrix of the passive network.

[0103] In this embodiment, the self-admittance element can be represented as a diagonal element extracted from a unified second-order complex vector converter admittance model, characterizing the converter's own port admittance. It reflects the fundamental impedance characteristics of the converter's output port when coupling with other frequency sequences is ignored.

[0104] Specifically, the self-admittance element can be an element characterizing the converter's own port admittance characteristics. This element can be calculated using the dynamic characteristics of the phase-locked loop and the parameters of the current control loop, specifically including the equivalent admittance frequency response characteristics of the converter in the αβ coordinate system. The addition operation can be implemented through matrix operations, superimposing the self-admittance element onto the complex vector node admittance matrix at the position corresponding to the converter's connected node. The main frequency block can be represented as an n×n submatrix in the upper left corner of the matrix, handling the dynamics near the fundamental frequency; the coupling frequency block can be represented as an n×n submatrix in the lower right corner, handling the coupling components after frequency shift. The position of the main diagonal is determined based on the node numbering rules. For example, when connected to node i, the position corresponding to the main frequency block is (i, i), and the position corresponding to the coupling frequency block is (n+i, n+i).

[0105] Add the self-admittance element from the output admittance model of the grid converter to the main diagonal position of the main frequency block and the coupling frequency block corresponding to the node connected to the grid converter in the complex vector node admittance matrix of the passive network.

[0106] In this embodiment, the integration of the self-admittance element of the grid-type converter can adopt the same matrix operation rules as described above, but its admittance characteristics differ. Specifically, the self-admittance element of the grid-type converter can be jointly determined by the power droop control characteristics (e.g., active power-frequency droop coefficient) and the voltage-current dual-loop control parameters, exhibiting voltage source impedance characteristics. In practice, its equivalent output admittance can be calculated according to the control mode of the grid-type converter (e.g., VSG mode or droop control mode). The addition location also follows the node correspondence principle, but the calculation of the admittance value needs to consider the influence of virtual inertial elements and power regulation dynamics, so that the admittance characteristics of the grid-type converter in the low-frequency range differ from those of the grid-type converter.

[0107] In the output admittance models of the grid-type converter and the network-type converter, the cross admittance elements representing the frequency coupling effect are added to the corresponding non-main diagonal positions in the complex vector node admittance matrix, respectively. The cross admittance elements include admittance functions corrected by frequency offset and rotation factor. Through the addition operation, the wideband coupling admittance network model is formed, and the block diagonal structure of the complex vector node admittance matrix is ​​maintained.

[0108] In this embodiment, the cross-admittance element can be represented as an off-diagonal admittance element extracted from the converter admittance model, specifically used for quantifying frequency coupling effects.

[0109] Specifically, cross-admittance elements can be represented as off-diagonal elements in the converter admittance matrix. These elements can be obtained by correcting for frequency offset functions and rotation factors, and are specifically used to characterize frequency coupling effects. During integration, cross-admittance elements can be added to the off-diagonal positions of the complex vector node admittance matrix, such as the intersection between the main frequency block and the coupled frequency block, the coupling positions between different converter access nodes, and the correlation positions between different frequency components of the same converter.

[0110] In this embodiment, firstly, by precisely adding the self-admittance element of the converter to the main diagonal positions of the main frequency block and the coupling frequency block of the complex vector node admittance matrix, the model ensures the complete preservation of the converter's dynamic characteristics. Simultaneously, by embedding the cross-admittance element characterizing the frequency coupling effect (e.g., the admittance function corrected by frequency offset and rotation factor) into the non-main diagonal positions of the matrix, the interaction mechanism of voltage-current-frequency over a wide frequency range is directly revealed, avoiding information loss due to coordinate transformation in traditional methods. Secondly, this integration operation strictly maintains the block diagonal structure of the matrix, which not only improves the computational efficiency and numerical stability of the model but also supports dynamic updates and scalability of the model when the system topology changes. This provides a high-precision, unified modeling foundation for real-time stability analysis and oscillation suppression strategies in new energy power plants.

[0111] This embodiment provides a method for updating a wideband coupled admittance network model. The execution subject of the method can be a field monitoring computer, a server, etc.

[0112] The method may include:

[0113] Step S22: Monitor topology change events in the power plant; wherein the topology change events include the commissioning or disconnection of converters or the switching on or off of electrical connections.

[0114] In this embodiment, the executing entity (station monitoring computer) can communicate with the power plant's monitoring and data acquisition system or relay protection system to receive position status signals (e.g., closed or open) from switching equipment such as circuit breakers, disconnectors, and contactors.

[0115] In this embodiment, the executing entity can also receive operating status words or control mode commands from the converter controller itself to determine whether the converter has been remotely connected or disconnected. The monitoring process can be either passively receiving status change signals or actively and periodically polling the status of each device.

[0116] In this embodiment, the commissioning of a converter can be represented as an event in which a converter that was not previously connected to the grid (whether GFL or GFM) is allowed to be connected to the AC bus of the power plant and begin transmitting power. The decommissioning of a converter can be represented as the disconnection of an operating converter from the power grid. The connection or disconnection of electrical links can be represented as a change in the state of the electrical path connecting various nodes within the power plant (e.g., collector lines, tie switches).

[0117] Step S24: In response to the detected topology change event, invoke the pre-stored broadband coupled admittance network model correction strategy.

[0118] In this embodiment, the pre-stored wideband coupled admittance network model correction strategy can be represented as a series of encapsulated matrix operation rules for specific topology change types.

[0119] Specifically, the pre-stored wideband coupled admittance network model correction strategy can be a configuration file. Specifically, it can be a rule base stored in XML, JSON, or YAML format, which defines matrix operation instructions corresponding to different event types.

[0120] The pre-stored wideband coupled admittance network model correction strategy can also be a database table or an embedded code module.

[0121] In a specific implementation, after step S22 identifies a specific event (e.g., GFL converter tripping or A collector line disconnection), the execution entity can index an internal event-policy mapping table based on the event type (e.g., node tripping or branch disconnection) to find the corresponding correction policy program entry address or function. Then, the execution entity can execute the corresponding correction algorithm code or load the corresponding computation module.

[0122] Step S26: Update the broadband coupled admittance network model of the power plant according to the correction strategy of the pre-stored broadband coupled admittance network model, so as to maintain its accurate representation of the dynamic characteristics of the power plant under the current operating conditions; wherein, the broadband coupled admittance network model is obtained by the above-mentioned modeling method of a multi-converter coupled system.

[0123] In this embodiment, if the strategy is to add a new node, the execution entity can dynamically allocate new memory space for the system admittance matrix, expanding it from 2n×2n to 2(n+1)×2(n+1), and initialize its admittance elements in the newly added rows and columns according to the electrical parameters and connection relationships of the new node.

[0124] If the strategy is to remove a node, the executing entity can determine the two rows and two columns (in the αβ coordinate system) corresponding to the node, delete it from the matrix as a whole, and compress the matrix dimension at the same time.

[0125] If the strategy is to disconnect the line, the execution entity can locate the specific off-diagonal element in the matrix that represents the mutual admittance between the two nodes connected by the line and modify it to zero; at the same time, according to Kirchhoff's current law, the self-admittance elements corresponding to these two nodes are adjusted synchronously (that is, the admittance value of the line is subtracted).

[0126] In some implementations, updating the broadband coupled admittance network model of the power plant according to the pre-stored broadband coupled admittance network model correction strategy includes:

[0127] Step S262: In response to the topology change event, perform a matrix correction operation corresponding to the event type; wherein, the matrix correction operation is configured as follows:

[0128] If the topology change event is the addition of a new converter or node, then in the complex vector node admittance matrix of the wideband coupled admittance network model, a row and column corresponding to the new converter or node are added, and the new admittance element is initialized based on its electrical parameters.

[0129] In this implementation, it is assumed that the original system has n physical nodes, corresponding to a 2n×2n complex vector node admittance matrix. When a new node is added, the execution entity can add two rows and two columns to the matrix in memory, making it 2(n+1)×2(n+1) dimensional. The positions of the newly added row and column indices can be uniquely determined by the system's node numbering rules.

[0130] For a newly added node, the elements at the diagonal positions in the main frequency block and coupled frequency block of the new matrix can be initialized as the sum of the admittances of all branches directly connected to that node.

[0131] If the newly added node is a converter, its self-admittance can be superimposed on the self-admittance elements in its own output admittance model. At the corresponding positions between the new node and existing nodes, their mutual admittance elements can be initialized to negative values ​​of the admittance of the lines or transformers connecting them. If the node is an isolated node, these mutual admittance elements are initialized to zero.

[0132] If the topology change event is the removal of a converter or node, then all rows and columns corresponding to the removed converter or node are deleted from the complex vector node admittance matrix.

[0133] In this implementation, the executing entity first determines the range of row and column indices occupied by the cut-off converter or node in the original 2n×2n matrix (one physical node corresponds to two rows and two columns). Then, a preset deletion algorithm is executed to completely remove all these rows and columns from the matrix, thereby generating a new 2(n-1)×2(n-1) dimensional matrix. This operation not only removes the node's own dynamics but also automatically disconnects it from all other nodes in the network.

[0134] If the topology change event is an electrical connection line disconnection, then the mutual admittance elements between the nodes connected to the disconnected line in the complex vector node admittance matrix are set to zero, and the self-admittance elements of the connected nodes are corrected simultaneously.

[0135] In this embodiment, assuming the line connecting node i and node j is disconnected, the execution entity can locate all off-diagonal elements in the matrix that represent the coupling relationship between node i and node j, specifically including the elements in the main frequency block and the coupling frequency block corresponding to positions (i, j) and (j, i), and set them to zero.

[0136] In this embodiment, the synchronous correction of the self-admittance elements of the connected nodes can be represented as follows: after the mutual admittance is set to zero, the self-admittances (i.e., the main diagonal elements) of nodes i and j must be adjusted accordingly to maintain the current balance of the nodes. The executing entity will subtract the original mutual admittance value with node j from the self-admittance of node i, and perform the same operation on node j.

[0137] This embodiment provides an impedance stability analysis method for new energy power plants. The execution subject of the method can be a station monitoring computer, a server, etc.

[0138] The method may include:

[0139] Step S32: Establish a wideband coupled admittance network model or obtain an updated wideband coupled admittance network model; wherein, the wideband coupled admittance network model is modeled using the above-mentioned modeling method for a multi-converter coupled system; wherein, the updated wideband coupled admittance network model is obtained by updating the wideband coupled admittance network model using the above-mentioned updating method.

[0140] Step S34: Determine the equivalent impedance characteristics of the power generation field based on the broadband coupled admittance network model.

[0141] In this embodiment, the user can first explicitly specify an observation node in the execution entity. This node can be set as a common connection point for power exchange between the power plant and the external power grid. This node has a unique index identifier in the complex vector node admittance matrix corresponding to the broadband coupled admittance network model.

[0142] Then, the executing entity can perform a pre-defined matrix operation on the aforementioned system-level complex vector node admittance matrix, namely, block inversion. Specifically, the entire matrix can be divided into different blocks according to the specified observation node and all other nodes in the network. By performing rigorous mathematical transformations and solving this block matrix, a simplified equivalent admittance matrix specific to that observation point can be derived. It completely encapsulates all dynamic admittance characteristics viewed from the observation point into the interior of the power plant.

[0143] Finally, by obtaining the inverse of the equivalent admittance matrix, the executing entity can obtain the equivalent impedance characteristics of the power generation field. This equivalent impedance characteristic can be a second-order complex vector matrix. In this matrix, the elements on the main diagonal represent the self-impedance characteristics of the power generation field at the positive and negative sequence frequency components, respectively; while the elements on the off-diagonal directly quantify and characterize the strength and dynamic relationship of the coupling between the positive and negative sequence frequency components.

[0144] Step S36: Perform stability criterion analysis on the equivalent impedance characteristics to assess the stability risk of the power plant over a wide frequency range.

[0145] In this embodiment, the system back-to-back matrix can first be constructed. This system back-to-back matrix can be formed by combining the equivalent impedance characteristics of the power plant with the equivalent impedance characteristics of the grid side, and its essence reflects the dynamic interaction strength between the power plant and the grid.

[0146] Then, the executing entity can perform a comprehensive frequency scan analysis of the system's hysteresis matrix within a preset wide frequency range. During this process, for each target frequency point within the frequency range, all eigenvalues ​​corresponding to the hysteresis matrix are calculated. These eigenvalues ​​constitute the key dynamic identifiers of the system at different frequencies.

[0147] Next, all the calculated eigenvalues ​​can be plotted on the complex plane to form a series of trajectory curves of eigenvalues ​​changing with frequency, namely the generalized Nyquist trajectory.

[0148] In this embodiment, stability can be determined based on the following criteria: if none of the eigenvalue trajectories encircle the critical point with coordinates of negative one zero on the complex plane, then the power plant can be determined to be stable over a wide frequency range under the current operating conditions; conversely, if any eigenvalue trajectory encircles the critical point, it indicates that the system is at risk of instability, and the oscillation mode at the corresponding frequency may diverge.

[0149] In some implementations, after establishing or obtaining the broadband coupled admittance network model, a verification step of the model is further included:

[0150] At a predetermined node in the power plant, a predetermined harmonic disturbance current signal is injected.

[0151] In this embodiment, a known and controllable electrical disturbance can be applied at a certain location in the power plant network to stimulate the system's broadband dynamic response, thereby providing comparative data from measured or high-precision simulations for model verification.

[0152] In this embodiment, the preset node can be a common connection point of the power plant or a key internal node near the converter cluster.

[0153] In this embodiment, the preset harmonic disturbance current signal can be a single-frequency sinusoidal harmonic, or a sweep signal or wideband pulse that includes multiple frequency components, to cover a wide frequency range from subsynchronization to supersynchronization.

[0154] Based on the broadband coupled admittance network model, the harmonic voltage response of one or more nodes in the power plant is calculated.

[0155] In this embodiment, the executing entity can substitute the injected harmonic disturbance current signal as a current source vector into the complex vector node voltage equation corresponding to the broadband coupled admittance network model. By solving this system of linear equations, the harmonic voltage response at all nodes of the power plant or a series of key nodes specified by the operator can be calculated under the model prediction.

[0156] The calculated harmonic voltage response is compared with the reference voltage response obtained through dynamic simulation or physical measurement.

[0157] In this embodiment, the reference voltage response can be obtained by using a detailed model based on an electromagnetic transient simulation program, or it can be obtained by directly measuring the voltage response waveforms of each node during actual power plant operation using voltage sensors and a data acquisition system while actually injecting disturbances.

[0158] In this embodiment, the executing entity can compare the amplitude and phase errors of each harmonic in the frequency domain, or calculate the error norm (root mean square error) between waveforms in the time domain.

[0159] The accuracy of the broadband coupled admittance network model is verified based on the comparison results. After the verification is passed, the stability criterion analysis of the equivalent impedance characteristics is performed to assess the stability risk of the power plant in a broadband range.

[0160] In this embodiment, the executing entity can compare the calculated error with a preset error threshold. If the errors of all interested frequencies or nodes are lower than the threshold, the model is deemed to have passed validation, meaning its accuracy is considered to meet the requirements of engineering analysis.

[0161] In one specific implementation, a modeling method for a hybrid multi-converter coupled system is provided.

[0162] Specifically, in recent years, with the rapid development of new energy sources such as wind power and photovoltaics, new energy power plants have occupied an increasingly higher proportion of the power system. Traditional power systems centered on synchronous generators are gradually being replaced by high-proportion power electronic interface systems, leading to a decrease in system inertia, weakened voltage support capacity, and significant changes in operating characteristics. Under high penetration conditions, the dynamic behavior of new energy power plants has become a key factor determining the stability of the power system.

[0163] In renewable energy power plants, generating units are connected to the power grid via power electronic converters. These converters can be broadly classified into two types: grid-linked converters (GFLs) and grid-connected converters (GFMs). GFLs track the grid voltage phase through phase-locked loops and rely on the external grid for frequency reference; their output characteristics are closer to those of a current source. GFMs, on the other hand, actively establish voltage and frequency references through mechanisms such as voltage controllers and droop control, exhibiting voltage source characteristics. In practical engineering, both types of converters are often connected to the same power plant in a hybrid configuration, forming a complex coupling system with the external grid.

[0164] When grid-connected and grid-connected converters operate together, the dynamic response of the system is affected by multiple control mechanisms. Grid-connected converters are sensitive to voltage disturbances, while grid-connected converters have a certain degree of resilience. When the capacity ratio or control parameters of the two are not configured properly, multimodal oscillations may occur over a wide frequency range. Especially in scenarios with weak grids or high proportions of renewable energy integration, hybrid systems may experience low-frequency, medium-frequency, and even high-frequency instability problems, posing a serious challenge to the safe operation of renewable energy power plants and the power system.

[0165] Current stability studies are mostly based on impedance modeling methods, which treat a single converter as an equivalent impedance source and perform interactive analysis in conjunction with grid impedance. However, this type of method has three shortcomings: multi-machine systems are often equivalent to single-machine models, which cannot reflect the interaction between different types of converters; existing impedance modeling is mostly aimed at GFL converters, and insufficient attention is paid to the active control characteristics of GFM converters; and the sequence impedance method relies on coordinate transformation and local phase information, making it difficult to uniformly reveal the frequency coupling effects in hybrid systems.

[0166] To address the aforementioned challenges, it is necessary to propose a unified modeling method: simultaneously establishing output impedance models for both grid-connected and grid-connected converters within a complex vector framework of the αβ coordinate system. This method not only clearly characterizes the coupling relationship between voltage, current, and frequency but also intuitively reveals the mechanism of interaction among multiple converters during coupled operation. Furthermore, by combining this method with the internal lines and passive networks of the power plant, a complex vector node admittance matrix for the entire system is constructed, thereby unifying the coupling effects of multiple converters into the system-level description. Based on this, when grid-connected or grid-connected converters are added or removed, or when their operating conditions are adjusted, the overall model can be flexibly updated according to the correction rules of the complex vector node admittance matrix, thus achieving unified modeling and dynamic adjustment of the hybrid system of the new energy power plant.

[0167] Depend on Figure 2 As can be seen, this implementation scheme provides an impedance modeling method for grid-type converters and network-type converters under a complex vector modeling framework, including complex vector modeling of lines and passive components.

[0168] In traditional modeling methods, the impedance model of a grid-type converter in the dq coordinate system is as follows:

[0169] (1)

[0170] In formula (1), Let be the small-signal disturbance vector of the AC output voltage of the mesh converter relative to its steady-state operating point in the rotating dq coordinate system. Let be the small-signal disturbance vector of the AC output current of the mesh converter relative to its steady-state operating point in the rotating dq coordinate system. The superscript * indicates the conjugate component of the complex number. - Let be the equivalent impedance of the grid converter in the dq coordinate system. These four parameters together form a 2×2 equivalent impedance matrix. and These can be the main diagonal elements of the matrix, which respectively characterize the converter's impedance characteristics along the d-axis and q-axis. and It can be a non-diagonal element of a matrix, which represents the cross-coupling impedance between the d-axis and the q-axis.

[0171] However, since the dq coordinate system is based on rotating coordinate transformation, its frequency decoupling capability is limited. When a phase-locked loop (PLL) is introduced, the model will exhibit frequency coupling effects, making it difficult to fully characterize the dynamic characteristics of the system over a wide frequency range. In particular, when multiple converters are operating in parallel, the model in the dq coordinate system is prone to losing frequency interaction information between different nodes, resulting in inaccurate stability assessments.

[0172] To improve upon the above phenomenon, this implementation scheme further proposes a complex vector impedance modeling method in the αβ coordinate system. The complex vector voltage and current relationship between the αβ and dq coordinate systems is as follows:

[0173] (2)

[0174] (3)

[0175] In formulas (2) and (3), It can represent the voltage vector in the αβ stationary coordinate system. This can represent the current vector in the αβ stationary coordinate system. The superscript * indicates the conjugate component of the complex number. Let be the rotation factor expressed using Euler's formula. For rotation angle, Let be a Laplace variable, representing the complex frequency. Angular frequency, For frequency offset, , , , Let be the elements of the impedance matrix in the dq coordinate system, which are functions of the complex frequency s. It is the positive-sequence self-impedance (the transfer function from positive-sequence current to positive-sequence voltage). This represents the coupling impedance from negative sequence to positive sequence. This represents the coupling impedance from positive sequence to negative sequence. It is a negative sequence self-impedance.

[0176] Substituting equations (2) and (3) into equation (1), and simultaneously converting its impedance model into a unified admittance model, we obtain the complex vector admittance model of the grid converter as follows:

[0177] (4)

[0178] In formula (4), Let be the small-signal disturbance variable of the current at the output port of the grid converter in the αβ coordinate system. Let αβ be the voltage small-signal disturbance variable at the output port of the grid converter in the αβ stationary coordinate system. To provide the positive-sequence self-admittance function of the grid-type converter, This is the coupling admittance function from negative to positive order. This is the coupling admittance function from positive to negative order. To determine the negative-sequence self-admittance function of a grid-type converter. It is a complex exponential twitch factor. This is the fundamental angular frequency of the power grid.

[0179] Similarly, the unified impedance model of a grid-type converter in the dq coordinate system is as follows:

[0180] (5)

[0181] In formula (5), Represented as a voltage disturbance vector, where, The voltage small-signal disturbance at the output port of the grid converter in the dq rotating coordinate system, i.e., the voltage disturbance vector, is used as the output of the model to reflect the voltage response characteristics of the GFM converter under disturbance. Let be the current disturbance vector, where In the dq coordinate system, the small-signal current disturbance at the output port of the grid converter, that is, the current disturbance vector, is used as the input of the model to excite the system to generate a voltage response. For impedance matrix elements, For positive sequence self-impedance, This represents the coupling impedance from negative sequence to positive sequence. This represents the coupling impedance from positive sequence to negative sequence. It is a negative sequence self-impedance.

[0182] Grid-type converters employ power droop and dual-loop voltage-current control, with dynamic coupling between the outer-loop power control and the inner-loop voltage-current control. This coupling effect introduces additional frequency coupling into the system, making it impossible to accurately characterize its wide-band dynamic characteristics. In scenarios involving strong power grids or multiple converters operating in parallel, this may lead to significant discrepancies between stability analysis results and actual conditions.

[0183] Furthermore, to overcome the above shortcomings, this implementation scheme adopts a complex vector impedance modeling method in the αβ coordinate system. The coordinate transformation process is similar to the principle described in equations (2) and (3). Based on this transformation, the complex vector admittance model of the grid-type converter in the αβ coordinate system can be obtained, and its specific expression is as follows:

[0184] (6)

[0185] In formula (6), Let be the small-signal disturbance vector of the output current of the GFM converter in the αβ coordinate system. To introduce a frequency-coupled current disturbance term, The positive-sequence self-admittance is the transfer function characterizing the positive-sequence voltage perturbation to the positive-sequence current response. This is the negative-sequence to positive-sequence coupling admittance, which is used to capture the effect of negative-sequence voltage disturbances on positive-sequence current. This is the positive-sequence to negative-sequence coupling admittance, which characterizes the ability of a positive-sequence voltage perturbation to induce a negative-sequence current response. This is the negative-sequence self-admittance, used to describe the self-admittance characteristics at negative-sequence frequencies. It is the voltage disturbance vector in the αβ coordinate system. It is a frequency-coupled voltage term.

[0186] In the proposed complex vector modeling method, the admittance model of the grid converter can reveal the frequency coupling characteristics caused by the phase-locked loop, thus making it suitable for stability analysis of multi-converter systems. Similarly, the admittance model of the grid converter can characterize the frequency coupling effect caused by power droop control and the dynamic action of voltage and current loops, thereby providing a reliable modeling basis for wideband stability assessment of multi-machine parallel systems.

[0187] Secondly, a model of passive components and impedance lines within the new energy power plant is constructed, and its complex vector admittance model is as follows:

[0188] (7)

[0189] In formula (7), This represents a small-signal perturbation in the frequency domain of the current vector flowing through the passive element between nodes m and n in the αβ stationary coordinate system. for The complex conjugate, but with a frequency shift. , Let be the voltage difference vector between nodes m and n in the αβ coordinate system. for The complex conjugate, with the same frequency shift. , Let m be the equivalent frequency domain admittance of the passive element between nodes m and n in the αβ coordinate system. The complex vector admittance model of the entire passive element is a 2×2 block diagonal matrix.

[0190] The complex vector admittance model extended to n nodes is as follows:

[0191] (8)

[0192] In formula (8) It is a complex matrix of size 2n × 2n, consisting of four n × n submatrix blocks. It represents the equivalent admittance characteristics of the entire power plant and integrates the admittance models of all passive network components (e.g., lines, transformers) and converters (grid-connected GFL and grid-connected GFM).

[0193] Let be an n×n complex matrix, representing the nodal admittance matrix at the principal frequency. The matrix elements are functions of the complex frequency s (i.e., transfer functions).

[0194] Let be an n×n complex matrix, representing the nodal admittance matrix at the coupling frequency. This is achieved through frequency shift. This is obtained and used to capture frequency coupling effects. It is an n×n matrix with all elements equal to zero. In a block diagonal structure, a zero matrix indicates that there is no direct coupling between the main frequency block and the coupled frequency block. Let be an n×1 complex vector representing the small-signal voltage perturbations at the dominant frequency of n nodes in the αβ coordinate system. Each element corresponds to the voltage perturbation (a combination of α and β components) of one node. Let be an n×1 complex vector, and let be the injected current vector, representing the small-signal current disturbance injected into n nodes in the αβ coordinate system. yes The conjugate variable represents the frequency shift. . for The conjugate, but with a frequency shift. .

[0195] Furthermore, taking a three-node system in a power plant as an example, a grid-connected converter and a network-connected converter are connected to the system for hybrid operation. Its node admittance matrix... Make the following modifications:

[0196] (9)

[0197] in, , ,

[0198]

[0199] ,

[0200] ,

[0201]

[0202] In formula (9), It can be represented as an initial complex vector node admittance matrix, which only contains the admittance characteristics of passive network elements (transmission lines, grid-side equivalent admittance) and has not yet integrated the converter model. This can be represented as a modified complex vector nodal admittance matrix, in Based on this, the admittance models of grid-connected converters and network-connected converters are integrated to form a complete wideband coupled admittance network model. and This represents the equivalent admittance of the line between nodes, where, It is the admittance of the line between node 1 and node 2. It is the admittance of the line between node 1 and node 3. The grid-side equivalent admittance represents the equivalent admittance at the connection point between the power plant and the grid (e.g., node 1), including the effects of grid impedance and components such as transformers. This is the positive-sequence self-admittance of the GFL, characterizing the transfer function from positive-sequence voltage to positive-sequence current. Its subscript indicates the specific GFL converter. This is the positive-sequence self-admittance of the GFM, which is affected by power droop control and virtual inertia. Its subscript indicates the specific grid-type converter (GFM). and All are rotation factors. and These are the phase angles of the GFL and GFM converters, respectively. In the middle, the block in the upper left corner is the main frequency block, and its elements are constructed based on Kirchhoff's laws. It is the self-admittance of node 1 (the sum of the admittances of all connecting lines and the power grid). To represent the mutual admittance between node 1 and node 2 (negative values ​​indicate current outflow), other locations are filled with line admittance, and the admittance between non-connected nodes is zero. In the middle, the lower right block is the coupling frequency block, whose structure is similar to the upper left block, but all admittance evaluations are performed in... Frequency point. At In Based on this, modifications are made by adding converter admittance elements. Specifically, in the main frequency block, the self-admittance of the GFL is adjusted. Add to position (2, 2) to make the self-admittance of node 2 become .

[0203] Based on Equation (9), it can be seen that the modeling method proposed in this implementation plan not only considers the equivalent output admittance model of the hybrid converter, but also incorporates the grid-side equivalent admittance into the complex vector node admittance model, thus forming a system model containing complete information.

[0204] (10)

[0205] In formula (10), Let be the value of the voltage small-signal response vector of all n nodes in the power plant in the frequency domain, in the αβ stationary coordinate system. for The complex conjugate vector, but with a frequency shift. This parameter is used to capture the frequency coupling effect in the voltage response, i.e., the negative-sequence or offset frequency response that a positive-sequence voltage disturbance may excite. Let be the value in the frequency domain of the small-signal current disturbance vector injected into each node of the power plant in the αβ coordinate system. This value is used as the input to the model to generate the voltage response. for The complex conjugate vector, frequency offset . It is the inverse of the admittance matrix, which serves as the core of the linear transformation, converting input current disturbances into output voltage responses.

[0206] To verify the correctness of the designed admittance matrix, a harmonic current was injected at node m. The voltage values ​​of each node are calculated according to equation (10), and the results are obtained by comparing them with the voltage responses of each node in the simulation. The correctness of the model. Specifically... Verification process and results are as follows Figure 3 and Figure 4 As shown. Among them, Figure 4 (a) shows the voltage amplitude distribution verification at node 1. Figure 4 (b) Verification of voltage phase distribution at node 1.

[0207] like Figure 5 and Figure 6 As can be seen, the node admittance matrix constructed in this embodiment has good versatility. When a new grid-connected device is connected to the system, whether it is a single converter or a unit composed of multiple converters, it is only necessary to fill the corresponding position of the matrix with the equivalent admittance of the corresponding converter, without changing the overall form of the admittance matrix.

[0208] like Figure 5 and Figure 6 As shown, in a new energy power station system with m nodes, the addition and deletion of nodes can be uniformly handled by adjusting the order of the node admittance matrix. When a new node is added, the matrix order is expanded from 2m to 2m+2, and the admittance relationship corresponding to the new node will be reflected in the i-th row and i-th column, and the 2i-th row and 2i-th column of the matrix, respectively, which can fully characterize the coupling effect between the new node and the original nodes. When a node is removed, only its corresponding rows and columns need to be deleted, and the matrix order is reduced to 2m-2, thereby achieving reduced-order modeling of the system. This method enables the dynamic updating of the system topology to be completed within a unified matrix framework, maintaining the integrity of the model while improving the flexibility of the analysis.

[0209] Furthermore, if only the branch between any two nodes in the system is broken, there is no need to delete rows and columns. Instead, the mutual admittance elements between the two nodes in the matrix are set to zero, and the self-admittance terms of the relevant nodes are corrected at the same time to ensure that the network topology after the disconnection can be accurately represented in the matrix.

[0210] The modeling method for hybrid multi-converters proposed in this application has the following characteristics:

[0211] (1) By establishing the complex vector node admittance matrix, the coupling effect of the hybrid multi-converter system can be fully reflected, providing theoretical support for ensuring the stable operation of the system.

[0212] (2) Compared with the scalar node admittance matrix in the dq coordinate system, the complex vector node admittance matrix in the αβ coordinate system is a sparse matrix structure, which can effectively analyze the stability of the hybrid multi-current converter.

[0213] (3) The complex vector node admittance model can be flexibly adjusted according to different operating conditions or topology changes (such as converter access, withdrawal or line disconnection) to ensure the stable operation of the system under various changes.

[0214] According to an embodiment of the present invention, an electronic device is provided. The electronic device in this embodiment may include one or more of the following components: a processor, a network interface, memory, non-volatile memory, and one or more application programs, wherein the one or more application programs may be stored in the non-volatile memory and configured to be executed by one or more processors, and the one or more programs are configured to perform the methods as described in the foregoing method embodiments.

[0215] According to embodiments of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a computer, causes the computer to perform the method described in any of the above embodiments.

[0216] According to embodiments of the present invention, a computer program product comprising instructions is also provided, which, when executed by a computer, cause the computer to perform a method in any of the above embodiments.

[0217] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A modeling method for a multi-converter coupled system, characterized in that, include: Obtain the electrical parameters of the passive network in the power plant, and the control parameters of the grid-connected converter and the grid-connected converter; Based on the electrical and control parameters, a broadband coupled admittance network model of the power plant is constructed in a first coordinate system, wherein the first coordinate system is a stationary coordinate system used to characterize the instantaneous characteristics of AC electrical quantities; wherein the broadband coupled admittance network model integrates the output admittance models of the grid-connected converter and the grid-connected converter with the admittance model of the passive network through a unified second-order complex vector structure, so as to directly characterize the voltage-current-frequency broadband coupling effect caused by converter control interaction.

2. The method according to claim 1, characterized in that, The unified second-order complex vector structure characterizes the frequency coupling effect in the following way: The output admittance model of the grid-connected converter is configured such that its output current disturbance vector is equal to the product of its own admittance function and the terminal voltage disturbance vector, plus a coupling term; wherein the coupling term is composed of the product of the frequency-shifted admittance function, the rotation factor, and the conjugate of the terminal voltage disturbance vector. The output admittance model of the grid converter is configured to be expressed in an isomorphic form consistent with the grid converter model, and its admittance function reflects the power droop and voltage-current dual-loop control characteristics of the grid converter.

3. The method according to claim 1, characterized in that, The admittance model of the passive network is represented by a complex vector node admittance matrix, which is configured as follows: For a network with n nodes, the complex vector node admittance matrix is ​​a 2n×2n block diagonal matrix. Its upper left n×n submatrix represents the admittance relationship of the node at the main frequency, and its lower right n×n submatrix represents the admittance relationship of the node at the coupling frequency after frequency shift.

4. The method according to any one of claims 1 to 3, characterized in that, The integration of the output admittance models of the grid-connected converter and the network-type converter with the admittance model of the passive network includes: Add the self-admittance element in the output admittance model of the grid converter to the main diagonal position of the main frequency block and the coupling frequency block corresponding to the node connected to the grid converter in the complex vector node admittance matrix of the passive network. Add the self-admittance element in the output admittance model of the grid converter to the main diagonal position of the main frequency block and the coupling frequency block corresponding to the node connected by the grid converter in the complex vector node admittance matrix of the passive network; In the output admittance models of the grid-type converter and the network-type converter, the cross admittance elements representing the frequency coupling effect are added to the corresponding non-main diagonal positions in the complex vector node admittance matrix, respectively. The cross admittance elements include admittance functions corrected by frequency offset and rotation factor. Through the addition operation, the wideband coupling admittance network model is formed, and the block diagonal structure of the complex vector node admittance matrix is ​​maintained.

5. A method for updating a broadband coupled admittance network model, characterized in that, include: Monitor topology change events in the power plant; wherein, the topology change events include the commissioning or disconnection of converters or the switching on or off of electrical connections; In response to detected topology change events, a pre-stored broadband coupled admittance network model correction strategy is invoked; According to the correction strategy of the pre-stored broadband coupled admittance network model, the broadband coupled admittance network model of the power plant is updated to maintain its accurate representation of the dynamic characteristics of the power plant under the current operating conditions; wherein, the broadband coupled admittance network model is obtained by modeling a multi-converter coupled system using any one of the modeling methods described in claims 1-4.

6. The updating method according to claim 5, characterized in that, The step of updating the broadband coupled admittance network model of the power plant according to the correction strategy based on the pre-stored broadband coupled admittance network model includes: In response to the topology change event, a matrix correction operation corresponding to the event type is performed; wherein the matrix correction operation is configured as follows: If the topology change event is the addition of a new converter or node, then in the complex vector node admittance matrix of the wideband coupled admittance network model, a row and column corresponding to the new converter or node are added, and the new admittance element is initialized based on its electrical parameters. If the topology change event is the removal of a converter or node, then all rows and columns corresponding to the removed converter or node are deleted from the complex vector node admittance matrix. If the topology change event is an electrical connection line disconnection, then the mutual admittance elements between the nodes connected to the disconnected line in the complex vector node admittance matrix are set to zero, and the self-admittance elements of the connected nodes are corrected simultaneously.

7. A method for impedance stability analysis of a new energy power plant, characterized in that, include: Establish a wideband coupled admittance network model or obtain an updated wideband coupled admittance network model; wherein, the wideband coupled admittance network model is constructed using the modeling method of a multi-converter coupled system as described in any one of claims 1-4; wherein, the updated wideband coupled admittance network model is obtained by updating the wideband coupled admittance network model using the updating method of a wideband coupled admittance network model as described in any one of claims 5-6; The equivalent impedance characteristics of the power generation field are determined based on the broadband coupled admittance network model. A stability criterion analysis was performed on the equivalent impedance characteristics to assess the stability risk of the power plant over a wide frequency range.

8. The impedance stability analysis method according to claim 7, characterized in that, After establishing or obtaining the broadband coupled admittance network model, the method further includes a verification step for the model: At a predetermined node in the power plant, a predetermined harmonic disturbance current signal is injected. Based on the broadband coupled admittance network model, the harmonic voltage response of one or more nodes in the power plant is calculated; The calculated harmonic voltage response is compared with the reference voltage response obtained through dynamic simulation or physical measurement. The accuracy of the broadband coupled admittance network model is verified based on the comparison results. After the verification is passed, the stability criterion analysis of the equivalent impedance characteristics is performed to assess the stability risk of the power plant in a broadband range.

9. An electronic device, characterized in that, include: A memory, and one or more processors communicatively connected to the memory; The memory stores instructions that can be executed by the one or more processors to cause the one or more processors to implement the method as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 4.

Images