Method and system for stability evaluation of offshore wind power flexible direct current grid-connected system based on multi-access point simulation

Abstract

This invention relates to the field of power system stability analysis technology, and discloses a method and system for stability assessment of offshore wind power flexible DC grid-connected systems based on multi-access point simulation. The method includes: constructing an electromagnetic transient simulation model integrating an aggregated equivalent model of the offshore wind farm, a detailed model of the flexible DC transmission system, and an equivalent model of the receiving-end power grid; synchronously and parallelly injecting power disturbance signals at multiple physical connection points to excite the system's broadband dynamic response; collecting voltage and current time-series data at each connection point; constructing a multi-port admittance matrix characterizing the dynamic coupling relationship of multiple ports through frequency domain transformation; performing frequency domain fitting using the vector matching method to construct a closed-loop state-space model and extract the dominant poles of the system; and comparing the real parts of the poles with the stability boundary threshold to output the assessment conclusion. This invention, through synchronous disturbance injection at multiple access points, fully considers the dynamic coupling characteristics between multiple access points, accurately identifies broadband instability risks in the system, and significantly improves assessment accuracy and engineering adaptability.

Description

Technical Field

[0001] This invention relates to the field of power system stability analysis technology, specifically to a method and system for stability assessment of offshore wind power flexible DC grid-connected systems based on multi-access point simulation. Background Technology

[0002] With the acceleration of the large-scale development of offshore wind power, the use of flexible DC transmission technology to achieve long-distance transmission of offshore wind power has become the mainstream technical solution. The offshore wind power flexible DC grid connection system includes multiple dynamic links such as wind turbines, submarine cables, converter valves, DC lines and receiving-end power grids. There are complex broadband dynamic interactions between these links, which poses a severe challenge to the stable operation of the system.

[0003] Currently, two main technical approaches are used for stability assessment of offshore wind power flexible DC grid-connected systems: one is the eigenvalue analysis method based on a single-unit infinite bus system, which treats the entire wind farm as a single wind turbine and injects disturbances at only one connection point for impedance measurement, failing to reflect the dynamic coupling characteristics between multiple wind farm clusters at different connection points; the other is the detailed model method based on time-domain simulation, which, while able to accurately simulate the system's dynamic response, involves enormous computational costs and struggles to reveal the system's instability modes and dominant factors at the mechanistic level. A common drawback of these existing technologies is that they do not fully consider the dynamic interaction between multiple offshore wind farm connection points, leading to biased stability assessment results, difficulty in accurately identifying potential instability risks across a wide frequency band, and inability to provide a reliable basis for operation control and parameter optimization in practical engineering. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for stability evaluation of offshore wind power flexible DC grid connection systems based on multi-access point simulation, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a stability evaluation method for offshore wind power flexible DC grid-connected systems based on multi-access point simulation, comprising the following steps: S1. Construct an electromagnetic transient simulation model for an offshore wind power flexible DC grid-connected system. This electromagnetic transient simulation model integrates an equivalent model of an offshore wind farm aggregation, a detailed model of a flexible DC transmission system, and an equivalent model of a receiving-end AC grid. S2. In the electromagnetic transient simulation model, locate multiple physical connection points between the offshore wind farm aggregated equivalent model and the detailed model of the flexible DC transmission system, and synchronously and in parallel inject power disturbance signals with preset amplitude and frequency at each physical connection point to stimulate the broadband dynamic response of the offshore wind power flexible DC grid connection system across the entire access point range. S3. During the process of injecting power disturbance signals, voltage timing response data and current timing response data at each physical connection point are collected simultaneously. S4. Perform frequency domain transformation on the collected voltage and current time-series response data, and construct a multi-port admittance matrix to characterize the dynamic interactive coupling relationship between each physical connection point based on the transformed frequency domain data. S5. Perform frequency domain fitting on the multi-port admittance matrix to construct a continuous domain closed-loop state space model of the offshore wind power flexible DC grid-connected system, and solve the eigenvalues ​​of the closed-loop state space model to extract the dominant poles that determine the dynamic characteristics of the system. S6. Compare the real part of the dominant pole of the system with the preset stability boundary threshold, and output the stability assessment conclusion of the offshore wind power flexible DC grid connection system under the current operating conditions based on the comparison result.

[0006] As a preferred technical solution of the present invention, the power disturbance signal injected in S2 is specifically a set of sinusoidal power disturbance signals with interleaved frequencies applied simultaneously at each physical connection point. The set of sinusoidal power disturbance signals with interleaved frequencies covers the wide frequency band to be analyzed, and the frequency difference between each sinusoidal signal is greater than a preset frequency resolution threshold.

[0007] As a preferred embodiment of the present invention, S4 includes: The voltage and current timing response data collected at each physical connection point are synchronized to eliminate the time delay introduced by the difference in data acquisition channels and obtain a synchronized time domain dataset. Perform a fast Fourier transform on the synchronized time-domain dataset to obtain the frequency domain voltage response and frequency domain current response values ​​at each frequency point corresponding to each physical connection point. Based on the frequency domain voltage response and frequency domain current response at each frequency point, the admittance matrix elements corresponding to each frequency point in the multi-port admittance matrix are constructed by solving a system of linear equations.

[0008] As a preferred embodiment of the present invention, the frequency domain fitting process performed on the multi-port admittance matrix in step S5 is specifically as follows: The vector matching method is used to approximate each matrix element in the multi-port admittance matrix by rational fraction combination, resulting in a set of frequency domain response characteristic expressions that share the same pole set; Based on the frequency domain response characteristic expression, the multi-port admittance matrix is ​​converted into an equivalent state space realization, and this state space realization is combined with the external network equations of the offshore wind power flexible DC grid-connected system to eliminate algebraic variables and construct a closed-loop state space model describing all dynamic characteristics of the system. The state matrix of the closed-loop state-space model is decomposed into eigenvalues ​​to obtain all eigenvalues. From all eigenvalues, several eigenvalues ​​with the largest real parts that reflect the overall stability trend of the system are selected as the dominant poles of the system.

[0009] As a preferred embodiment of the present invention, S6 includes: A real part stability boundary threshold is preset, which is a negative number or zero; Extract the real part value of each pole in the dominant poles of the system one by one, and compare each real part value with the real part stability boundary threshold; When the real part values ​​of all dominant poles of the system are less than the real part stability boundary threshold, an evaluation result is generated to characterize the dynamic stability of the offshore wind power flexible DC grid-connected system under the current operating conditions. When the real part of at least one dominant pole of the system is greater than or equal to the real part stability boundary threshold, an assessment result is generated that characterizes the risk of instability of the offshore wind power flexible DC grid-connected system under the current operating conditions.

[0010] A stability evaluation system for offshore wind power flexible DC grid connection systems based on multi-access point simulation, used to implement any one of the methods described above, includes: The model building unit is used to build an electromagnetic transient simulation model of an offshore wind power flexible DC grid-connected system. This electromagnetic transient simulation model integrates an equivalent model of an offshore wind farm aggregation, a detailed model of a flexible DC transmission system, and an equivalent model of a receiving-end AC grid. The disturbance injection unit is used to locate multiple physical connection points between the aggregated equivalent model of offshore wind farm and the detailed model of flexible DC transmission system in the electromagnetic transient simulation model, and to synchronously and in parallel inject power disturbance signals with preset amplitude and frequency at each physical connection point to stimulate the broadband dynamic response of the offshore wind power flexible DC grid-connected system across the entire access point range. The data acquisition unit is used to simultaneously acquire voltage timing response data and current timing response data at each physical connection point during the injection of power disturbance signals. The matrix construction unit is used to receive voltage timing response data and current timing response data, and to perform frequency domain transformation processing on the received data. Based on the transformed frequency domain data, a multi-port admittance matrix is ​​constructed to characterize the dynamic interactive coupling relationship between each physical connection point. The pole solving unit is used to receive the multi-port admittance matrix, perform frequency domain fitting on the multi-port admittance matrix, construct a continuous domain closed-loop state space model of the offshore wind power flexible DC grid-connected system, solve the eigenvalues ​​of the closed-loop state space model, and extract the dominant system poles that determine the dynamic characteristics of the system. The evaluation conclusion output unit is used to receive the dominant pole of the system, compare the real part of the dominant pole with the preset stability boundary threshold, and output the stability evaluation conclusion of the offshore wind power flexible DC grid connection system under the current operating conditions based on the comparison result.

[0011] As a preferred embodiment of the present invention, the power disturbance signal injected by the disturbance injection unit is specifically a set of sinusoidal power disturbance signals with interleaved frequencies applied simultaneously at each physical connection point. The set of sinusoidal power disturbance signals with interleaved frequencies covers the wide frequency band to be analyzed, and the frequency difference between each sinusoidal signal is greater than a preset frequency resolution threshold.

[0012] As a preferred embodiment of the present invention, the matrix construction unit includes: The data preprocessing subunit is used to synchronize the voltage timing response data and current timing response data at each physical connection point, eliminate the time delay introduced by the difference in data acquisition channels, and obtain the synchronized time domain dataset. The frequency domain transformation subunit is used to perform a fast Fourier transform on the synchronized time domain dataset to obtain the frequency domain voltage response value and frequency domain current response value at each frequency point at each physical connection point. The matrix element solving sub-unit is used to construct the admittance matrix element corresponding to each frequency point in the multi-port admittance matrix by solving a system of linear equations based on the frequency domain voltage response value and frequency domain current response value at each frequency point.

[0013] As a preferred embodiment of the present invention, the pole solving unit includes: The fitting approximation sub-unit is used to approximate each matrix element in the multi-port admittance matrix by rational fractional combination using the vector matching method, so as to obtain a set of frequency domain response characteristic expressions that share the same pole set; The model construction subunit is used to convert the multi-port admittance matrix into an equivalent state-space realization based on the set of frequency domain response characteristic expressions, and to combine the state-space realization with the external network equations of the offshore wind power flexible DC grid-connected system to eliminate algebraic variables and construct a closed-loop state-space model describing all dynamic characteristics of the system. The feature extraction subunit is used to perform eigenvalue decomposition on the state matrix of the closed-loop state-space model to obtain all eigenvalues, and selects several eigenvalues ​​with the largest real parts that reflect the overall stability trend of the system as the dominant poles of the system.

[0014] As a preferred embodiment of the present invention, the evaluation conclusion output unit includes: The threshold preset subunit is used to set a real part stability boundary threshold, which is a negative number or zero. The comparison sub-unit is used to extract the real part value of each pole in the dominant poles of the system one by one, and compare each real part value with the real part stability boundary threshold. The conclusion generation sub-unit is used to generate an assessment result characterizing the dynamic stability of the offshore wind power flexible DC grid-connected system under the current operating conditions when the real part values ​​of all dominant poles of the system are less than the real part stability boundary threshold, based on the comparison results of the comparison sub-unit. When the real part value of at least one dominant pole of the system is greater than or equal to the real part stability boundary threshold, an assessment result characterizing the risk of instability of the offshore wind power flexible DC grid-connected system under the current operating conditions is generated.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention synchronously and in parallel injects power disturbance signals at multiple physical connection points between the aggregated equivalent model of an offshore wind farm and the detailed model of a flexible DC transmission system. This can simultaneously excite the system's wideband dynamic response across the entire access point range, fully considering the dynamic coupling characteristics between multiple access points. It overcomes the shortcomings of traditional single-point measurement methods that cannot reflect the interactive effects of multiple ports, enabling the constructed multi-port admittance matrix to truly characterize the overall dynamic characteristics of the system and significantly improving the accuracy of stability assessment.

[0016] 2. This invention uses the vector matching method to approximate the multi-port admittance matrix with rational fractional combination, and constructs a closed-loop state-space model to solve for eigenvalues. The stability state is judged from the distribution position of the dominant poles of the system, realizing an accurate mapping from frequency domain response data to system dynamic characteristics. While ensuring the evaluation accuracy, this method significantly reduces the computational resource consumption required by time domain simulation methods, can quickly identify the instability mode and dominant factors of the system, and provide clear stability warning information for operators.

[0017] 3. This invention generates an evaluation conclusion by comparing the real part of the dominant pole of the system with a preset stability boundary threshold, forming a complete technical chain from disturbance injection, data acquisition, matrix construction, pole solving to conclusion output. The steps are logically rigorous and closely connected. It is not only applicable to conventional scenarios of symmetrical unipolar structures, but also adaptable to complex engineering scenarios such as true bipolar structures, modular multilevel converter topologies, and switching between different operating modes by adjusting the disturbance injection strategy and threshold setting method. It has good engineering adaptability and promotion and application value.

[0018] 4. This invention conducts stability assessments based on electromagnetic transient simulation models. It can compare multiple schemes for proposed offshore wind power bases during the planning and design phase, optimize the grid connection layout and control parameters of wind farm clusters, and also conduct periodic assessments of in-service systems during the operation phase to promptly identify potential subsynchronous oscillation risks, guide the deployment and deactivation of additional damping control measures and adjustments to operating modes, prevent broadband oscillation accidents from the source, and ensure the safe and stable operation of offshore wind power flexible DC grid connection systems. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall process of the stability assessment method for offshore wind power flexible DC grid-connected systems based on multi-access point simulation of the present invention. Figure 2 This is a structural framework diagram of the offshore wind power flexible DC grid connection system stability evaluation system based on multi-access point simulation of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1

[0022] A stability assessment method for offshore wind power flexible DC grid-connected systems based on multi-access point simulation includes the following steps: S1. Construct an electromagnetic transient simulation model for an offshore wind power flexible DC grid-connected system. This electromagnetic transient simulation model integrates an equivalent model of an offshore wind farm aggregation, a detailed model of a flexible DC transmission system, and an equivalent model of a receiving-end AC grid. S2. In the electromagnetic transient simulation model, locate multiple physical connection points between the offshore wind farm aggregated equivalent model and the detailed model of the flexible DC transmission system, and synchronously and in parallel inject power disturbance signals with preset amplitude and frequency at each physical connection point to stimulate the broadband dynamic response of the offshore wind power flexible DC grid connection system across the entire access point range. S3. During the process of injecting power disturbance signals, voltage timing response data and current timing response data at each physical connection point are collected simultaneously. S4. Perform frequency domain transformation on the collected voltage and current time-series response data, and construct a multi-port admittance matrix to characterize the dynamic interactive coupling relationship between each physical connection point based on the transformed frequency domain data. S5. Perform frequency domain fitting on the multi-port admittance matrix to construct a continuous domain closed-loop state space model of the offshore wind power flexible DC grid-connected system, and solve the eigenvalues ​​of the closed-loop state space model to extract the dominant poles that determine the dynamic characteristics of the system. S6. Compare the real part of the dominant pole of the system with the preset stability boundary threshold, and output the stability assessment conclusion of the offshore wind power flexible DC grid connection system under the current operating conditions based on the comparison result.

[0023] Furthermore, the power disturbance signal injected in S2 is specifically a set of sinusoidal power disturbance signals with interleaved frequencies applied simultaneously at each physical connection point. This set of interleaved sinusoidal power disturbance signals covers the wide frequency band to be analyzed, and the frequency difference between each sinusoidal signal is greater than the preset frequency resolution threshold.

[0024] Furthermore, S4 includes: The voltage and current timing response data collected at each physical connection point are synchronized to eliminate the time delay introduced by the difference in data acquisition channels and obtain a synchronized time domain dataset. Perform a fast Fourier transform on the synchronized time-domain dataset to obtain the frequency domain voltage response and frequency domain current response values ​​at each frequency point corresponding to each physical connection point. Based on the frequency domain voltage response and frequency domain current response at each frequency point, the admittance matrix elements corresponding to each frequency point in the multi-port admittance matrix are constructed by solving a system of linear equations.

[0025] Furthermore, in S5, the multi-port admittance matrix is ​​subjected to frequency domain fitting, specifically as follows: The vector matching method is used to approximate each matrix element in the multi-port admittance matrix by rational fraction combination, resulting in a set of frequency domain response characteristic expressions that share the same pole set; Based on the frequency domain response characteristic expression, the multi-port admittance matrix is ​​converted into an equivalent state space realization, and this state space realization is combined with the external network equations of the offshore wind power flexible DC grid-connected system to eliminate algebraic variables and construct a closed-loop state space model describing all dynamic characteristics of the system. The state matrix of the closed-loop state-space model is decomposed into eigenvalues ​​to obtain all eigenvalues. From all eigenvalues, several eigenvalues ​​with the largest real parts that reflect the overall stability trend of the system are selected as the dominant poles of the system.

[0026] Furthermore, S6 includes: A real part stability boundary threshold is preset, which is a negative number or zero; Extract the real part value of each pole in the dominant poles of the system one by one, and compare each real part value with the real part stability boundary threshold; When the real part values ​​of all dominant poles of the system are less than the real part stability boundary threshold, an evaluation result is generated to characterize the dynamic stability of the offshore wind power flexible DC grid-connected system under the current operating conditions. When the real part of at least one dominant pole of the system is greater than or equal to the real part stability boundary threshold, an assessment result is generated that characterizes the risk of instability of the offshore wind power flexible DC grid-connected system under the current operating conditions.

[0027] A stability evaluation system for offshore wind power flexible DC grid connection systems based on multi-access point simulation, comprising methods for achieving any of the above, including: The model building unit is used to build an electromagnetic transient simulation model of an offshore wind power flexible DC grid-connected system. This electromagnetic transient simulation model integrates an equivalent model of an offshore wind farm aggregation, a detailed model of a flexible DC transmission system, and an equivalent model of a receiving-end AC grid. The disturbance injection unit is used to locate multiple physical connection points between the aggregated equivalent model of offshore wind farm and the detailed model of flexible DC transmission system in the electromagnetic transient simulation model, and to synchronously and in parallel inject power disturbance signals with preset amplitude and frequency at each physical connection point to stimulate the broadband dynamic response of the offshore wind power flexible DC grid-connected system across the entire access point range. The data acquisition unit is used to simultaneously acquire voltage timing response data and current timing response data at each physical connection point during the injection of power disturbance signals. The matrix construction unit is used to receive voltage timing response data and current timing response data, and to perform frequency domain transformation processing on the received data. Based on the transformed frequency domain data, a multi-port admittance matrix is ​​constructed to characterize the dynamic interactive coupling relationship between each physical connection point. The pole solving unit is used to receive the multi-port admittance matrix, perform frequency domain fitting on the multi-port admittance matrix, construct a continuous domain closed-loop state space model of the offshore wind power flexible DC grid-connected system, solve the eigenvalues ​​of the closed-loop state space model, and extract the dominant system poles that determine the dynamic characteristics of the system. The evaluation conclusion output unit is used to receive the dominant pole of the system, compare the real part of the dominant pole with the preset stability boundary threshold, and output the stability evaluation conclusion of the offshore wind power flexible DC grid connection system under the current operating conditions based on the comparison result.

[0028] Furthermore, the power disturbance signal injected by the disturbance injection unit is specifically a set of sinusoidal power disturbance signals with interleaved frequencies applied simultaneously at each physical connection point. This set of sinusoidal power disturbance signals with interleaved frequencies covers the wide frequency band to be analyzed, and the frequency difference between each sinusoidal signal is greater than the preset frequency resolution threshold.

[0029] Furthermore, the matrix building units include: The data preprocessing subunit is used to synchronize the voltage timing response data and current timing response data at each physical connection point, eliminate the time delay introduced by the difference in data acquisition channels, and obtain the synchronized time domain dataset. The frequency domain transformation subunit is used to perform a fast Fourier transform on the synchronized time domain dataset to obtain the frequency domain voltage response value and frequency domain current response value at each frequency point at each physical connection point. The matrix element solving sub-unit is used to construct the admittance matrix element corresponding to each frequency point in the multi-port admittance matrix by solving a system of linear equations based on the frequency domain voltage response value and frequency domain current response value at each frequency point.

[0030] Furthermore, the pole solving unit includes: The fitting approximation sub-unit is used to approximate each matrix element in the multi-port admittance matrix by rational fractional combination using the vector matching method, so as to obtain a set of frequency domain response characteristic expressions that share the same pole set; The model construction subunit is used to convert the multi-port admittance matrix into an equivalent state-space realization based on the set of frequency domain response characteristic expressions, and to combine the state-space realization with the external network equations of the offshore wind power flexible DC grid-connected system to eliminate algebraic variables and construct a closed-loop state-space model describing all dynamic characteristics of the system. The feature extraction subunit is used to perform eigenvalue decomposition on the state matrix of the closed-loop state-space model to obtain all eigenvalues, and selects several eigenvalues ​​with the largest real parts that reflect the overall stability trend of the system as the dominant poles of the system.

[0031] Furthermore, the evaluation conclusion output unit includes: The threshold preset subunit is used to set a real part stability boundary threshold, which is a negative number or zero. The comparison sub-unit is used to extract the real part value of each pole in the dominant poles of the system one by one, and compare each real part value with the real part stability boundary threshold. The conclusion generation sub-unit is used to generate an assessment result characterizing the dynamic stability of the offshore wind power flexible DC grid-connected system under the current operating conditions when the real part values ​​of all dominant poles of the system are less than the real part stability boundary threshold, based on the comparison results of the comparison sub-unit. When the real part value of at least one dominant pole of the system is greater than or equal to the real part stability boundary threshold, an assessment result characterizing the risk of instability of the offshore wind power flexible DC grid-connected system under the current operating conditions is generated.

[0032] Example 2

[0033] Based on Example 1 above, this embodiment further refines the implementation process of the stability assessment method for offshore wind power flexible DC grid connection system based on multi-access point simulation in specific application scenarios. In particular, for wind farm cluster access scenarios in different marine environments, the construction of the multi-port admittance matrix and the extraction process of the system's dominant poles are described in detail.

[0034] Taking a large-scale offshore wind power base in my country's coastal area as an example, the base includes four offshore wind farm clusters, which are connected to an offshore converter station via submarine cables and then connected to the onshore receiving-end power grid via a flexible DC transmission system. In this embodiment, step S1 in embodiment 1 is first executed to construct an electromagnetic transient simulation model that includes an aggregated equivalent model of the four offshore wind farms, a detailed model of a symmetrical unipolar flexible DC transmission system, and an equivalent model of the receiving-end AC power grid. The aggregated equivalent model of the offshore wind farms is constructed using the single-machine multiplication method. The detailed model of the flexible DC transmission system includes converter valves, bridge arm reactors, DC cables, and the converter station control system. The equivalent model of the receiving-end AC power grid is represented using Thevenin equivalent circuits.

[0035] Then, step S2 is executed to locate the four physical connection points between the aggregated equivalent model of the offshore wind farm and the detailed model of the flexible DC transmission system, namely the grid connection points of the four offshore booster stations. At these physical connection points, the disturbance injection unit synchronously and in parallel injects a set of sinusoidal power disturbance signals with interleaved frequencies. The frequency range of this set of signals is set to 1Hz to 1000Hz, covering the subsynchronous oscillation and harmonic resonance frequency bands that may occur in the offshore wind power flexible DC grid connection system. The frequency difference between each sinusoidal signal is set to 0.5Hz, which is greater than the preset frequency resolution threshold of 0.2Hz, ensuring that the response characteristics of each frequency point can be clearly distinguished in the subsequent frequency domain analysis.

[0036] During the execution of S3, the data acquisition unit synchronously acquires three-phase voltage timing response data and three-phase current timing response data at four physical connection points at a sampling frequency of 20kHz. Considering that the difference in submarine cable length may cause signal transmission delay, the acquired data is first sent to the data preprocessing subunit for synchronization processing. This subunit performs time shift correction on the timing data based on the delay calibration value between each physical connection point and the simulation master clock, eliminating the time delay introduced by the difference in data transmission path, and obtaining the synchronized time domain dataset.

[0037] Next, step S4 is executed. The frequency domain transformation subunit applies a Hanning window function to the synchronized time domain dataset to suppress spectral leakage, and then performs a fast Fourier transform to obtain the frequency domain voltage response and frequency domain current response values ​​at each physical connection point within the range of 1Hz to 1000Hz. The matrix element solving subunit receives these frequency domain data and constructs a system of linear equations for each frequency point based on multiport network theory. For example, at frequency point f, the voltage and current relationship of the four ports satisfies Y(f)*V(f)=I(f), where Y(f) is the 4×4 multiport admittance matrix to be determined. By obtaining frequency domain data through multiple measurements under different disturbance modes, the matrix element solving subunit uses the least squares method to solve the overdetermined linear equation system to obtain all 16 admittance matrix elements corresponding to frequency point f in the multiport admittance matrix. The above process is repeated to traverse all frequency points, and finally a multiport admittance matrix covering the entire analysis frequency band is constructed.

[0038] Then, step S5 is executed, fitting and approximating the multi-port admittance matrix received by the sub-unit. A vector matching method is used to approximate each element of the matrix using rational fraction combinations. Taking element Y11 as an example, its frequency domain response data is fitted into the form of a sum of several partial fractions, ensuring that the fitting process for all matrix elements shares the same set of poles. This shared pole set approximation method guarantees that the fitting results accurately reflect the inherent consistency of the dynamic characteristics between each port. The model building sub-unit, based on the obtained rational fraction combination, converts it into an equivalent state space implementation, obtaining a set describing the dynamic characteristics of the four grid connection points. The state equations and output equations are combined with the control system equations of the flexible DC transmission system and the equivalent circuit equations of the receiving-end AC grid to eliminate intermediate algebraic variables. Finally, a closed-loop state-space model that can completely describe all the dynamic characteristics of the offshore wind power flexible DC grid-connected system is constructed. The feature extraction sub-unit performs QR eigenvalue decomposition on the state matrix of the closed-loop state-space model to obtain all the system eigenvalues. From these eigenvalues, five eigenvalues ​​with the largest real part and imaginary part in the range of 1Hz to 1000Hz are selected as the dominant poles of the system that determine the dynamic stability of the system.

[0039] Finally, step S6 is executed. The threshold preset subunit sets the real part stability boundary threshold to -0.5. The comparison subunit extracts the real part values ​​of the five dominant poles of the system one by one, which are -0.8, -1.2, -0.3, -2.1 and -1.5 respectively. The conclusion generation subunit compares these real part values ​​with the stability boundary threshold of -0.5 and finds that the real part value of one pole is -0.3, which is greater than the stability boundary threshold of -0.5. Based on this, the conclusion generation subunit generates an assessment result characterizing the offshore wind power flexible DC grid-connected system as having an instability risk under the current operating conditions. The assessment result suggests to the operators that under the current condition of four wind farm clusters operating at full capacity and the receiving end grid being in a weak AC system state, the system has a risk of subsynchronous oscillation, and additional damping control measures or adjustments to wind power output distribution need to be taken. This embodiment clearly demonstrates the application process of this method in a real complex scenario and its technical effect of identifying potential instability risks through a logically rigorous step derivation.

[0040] Example 3

[0041] This embodiment focuses on illustrating the adaptability of the offshore wind power flexible DC grid connection system stability evaluation system based on multi-access point simulation under different topologies and control strategies, and details the collaborative working mechanism of each unit of the system when dealing with parameter uncertainties and switching between multiple operating modes, so as to demonstrate the technical integrity and engineering practicality of the present invention.

[0042] Suppose we need to evaluate another offshore wind power planning project. This project adopts a modular multilevel converter topology, and the wind farm is connected to the flexible DC system after being collected by a diode rectifier unit. Unlike the symmetrical monopole structure in Example 2, the flexible DC system in this example adopts a true bipole structure, which includes three physical connection points: two converter transformer grid-side connection points and one neutral bus connection point. To verify the applicability of the system of the present invention to this topology, the following implementation process is carried out.

[0043] The model building unit first constructs an electromagnetic transient simulation model adapted to the true bipolar structure. This model integrates three equivalent models of offshore wind farms, each model corresponding to the sending end of one pole. The detailed model of the flexible DC transmission system includes modular multilevel converter valves of the upper and lower arms, arm reactors, true bipolar DC lines, and corresponding valve control and pole control systems. In this embodiment, the equivalent model of the receiving-end AC grid adopts a more refined IEEE 39-bus system standard model to simulate the characteristics of a strong receiving-end grid.

[0044] The disturbance injection unit locates three physical connection points and sets two different disturbance injection strategies based on the symmetry characteristics of the true bipolar system. The first strategy injects common-mode power disturbance signals at all three connection points simultaneously to analyze the overall dynamic response of the system. The second strategy injects differential-mode power disturbance signals only at the positive and negative connection points, without injecting at the neutral bus point, to analyze the dynamic coupling characteristics between poles. Both disturbance signals are multi-sine signals with interleaved frequencies, and the frequency range is extended to 2000Hz to cover the higher frequency band corresponding to the internal dynamic characteristics of the modular multilevel converter.

[0045] During the two sets of disturbance injection processes, the data acquisition unit synchronously acquires voltage timing response data and current timing response data at three physical connection points at a sampling frequency of 50kHz. The acquired data are labeled as common-mode disturbance dataset and differential-mode disturbance dataset, respectively.

[0046] The matrix construction unit processes the two datasets separately. The data preprocessing subunit first synchronizes the time-series data in each dataset to eliminate the inherent delay between sampling channels. The frequency domain transformation subunit performs a fast Fourier transform on the synchronized data to obtain frequency domain response data under common-mode and differential-mode excitation. The matrix element solving subunit combines the response data under common-mode and differential-mode excitation based on the superposition principle to decouple the frequency domain response characteristics of the three ports under positive-sequence, negative-sequence, and zero-sequence components. Based on the decoupled sequence component data, a system of linear equations is solved to construct a multi-port admittance matrix in the sequence component coordinate system. This method of constructing the admittance matrix in the sequence component coordinate system can more clearly reveal the dynamic coupling relationship of different sequence component channels in a true bipolar system, providing a more refined mathematical model for subsequent stability analysis.

[0047] The pole solving unit receives the order component multi-port admittance matrix. The fitting approximation subunit uses the vector matching method to approximate each element in the admittance matrix using rational fractional combination. Considering the internal harmonic characteristics of the modular multilevel converter, the fitting order is set to 30 to ensure accurate approximation of the high-frequency band. Based on the fitting results, the model building subunit converts the model into a state-space implementation and establishes it in conjunction with the detailed model of the valve control system, the logic model of the pole control system, and the IEEE 39-node system model. During the in conjunction process, the model building subunit specifically addresses the algebraic loop problem caused by the neutral bus potential fluctuation in the true bipolar system. By introducing a small time constant inertial element, the algebraic loop is broken, and a solvable closed-loop state-space model is finally constructed. The feature extraction subunit performs eigenvalue decomposition on the model and extracts the top ten eigenvalues ​​with the largest real parts as the dominant poles of the system.

[0048] The evaluation conclusion output unit receives these dominant poles of the system. The threshold preset subunit sets different stability boundary thresholds for different operating modes. For example, in the unipolar operating mode, the stability boundary threshold is set to -1.0; in the bipolar symmetrical operating mode, it is set to -0.8; and in the bipolar asymmetrical operating mode, it is set to -1.2. The comparison subunit selects the corresponding stability boundary threshold according to the operating mode set in the current simulation and compares the real part of each dominant pole of the system with it. The conclusion generation subunit generates the evaluation conclusion based on the comparison results. For example, in the bipolar asymmetrical operating mode and when a key line of the receiving-end power grid is out of service, it is found that the real part of a dominant pole is -0.9, while the threshold under the current mode is -1.2. The real part is greater than the threshold. Based on this, the conclusion generation subunit generates a stability warning conclusion and clearly points out that the instability risk is related to the weakening of the receiving-end power grid structure and the power imbalance between poles.

[0049] This embodiment, through a detailed description of a series of technical means such as true bipolar topology, order component decoupling analysis, and adaptive threshold setting for multiple operating modes, logically deduces that the system of the present invention can adapt to complex and ever-changing engineering realities and effectively handle stability assessment problems under different topologies and control strategies. The entire implementation process fully reflects the collaborative working logic between the various units of the system, that is, the complete technical chain from model construction, disturbance injection, data acquisition, matrix construction, pole solving to conclusion output, ensuring the integrity and feasibility of the technical solution. The technical effect of this embodiment is that, through rigorous logical deduction rather than data comparison, it proves that the system of the present invention has the adaptability and technical reliability to cope with various complex scenarios in actual engineering.

[0050] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A stability evaluation method for offshore wind power flexible DC grid-connected systems based on multi-access point simulation, characterized in that, Includes the following steps: S1. Construct an electromagnetic transient simulation model for an offshore wind power flexible DC grid-connected system. This electromagnetic transient simulation model integrates an equivalent model of an offshore wind farm aggregation, a detailed model of a flexible DC transmission system, and an equivalent model of a receiving-end AC grid. S2. In the electromagnetic transient simulation model, locate multiple physical connection points between the offshore wind farm aggregated equivalent model and the detailed model of the flexible DC transmission system, and synchronously and in parallel inject power disturbance signals with preset amplitude and frequency at each physical connection point to stimulate the broadband dynamic response of the offshore wind power flexible DC grid connection system across the entire access point range. S3. During the process of injecting power disturbance signals, voltage timing response data and current timing response data at each physical connection point are collected simultaneously. S4. Perform frequency domain transformation on the collected voltage and current time-series response data, and construct a multi-port admittance matrix to characterize the dynamic interactive coupling relationship between each physical connection point based on the transformed frequency domain data. S5. Perform frequency domain fitting on the multi-port admittance matrix to construct a continuous domain closed-loop state space model of the offshore wind power flexible DC grid-connected system, and solve the eigenvalues ​​of the closed-loop state space model to extract the dominant poles that determine the dynamic characteristics of the system. S6. Compare the real part of the dominant pole of the system with the preset stability boundary threshold, and output the stability assessment conclusion of the offshore wind power flexible DC grid connection system under the current operating conditions based on the comparison result.

2. The method according to claim 1, characterized in that, The power disturbance signal injected in S2 is specifically a set of sinusoidal power disturbance signals with interleaved frequencies applied simultaneously at each physical connection point. This set of interleaved sinusoidal power disturbance signals covers the wide frequency band to be analyzed, and the frequency difference between each sinusoidal signal is greater than the preset frequency resolution threshold.

3. The method according to claim 1, characterized in that, S4 includes: The voltage and current timing response data collected at each physical connection point are synchronized to eliminate the time delay introduced by the difference in data acquisition channels and obtain a synchronized time domain dataset. Perform a fast Fourier transform on the synchronized time-domain dataset to obtain the frequency domain voltage response and frequency domain current response values ​​at each frequency point corresponding to each physical connection point. Based on the frequency domain voltage response and frequency domain current response at each frequency point, the admittance matrix elements corresponding to each frequency point in the multi-port admittance matrix are constructed by solving a system of linear equations.

4. The method according to claim 1, characterized in that, In step S5, the multi-port admittance matrix undergoes frequency domain fitting, specifically as follows: The vector matching method is used to approximate each matrix element in the multi-port admittance matrix by rational fraction combination, resulting in a set of frequency domain response characteristic expressions that share the same pole set; Based on the frequency domain response characteristic expression, the multi-port admittance matrix is ​​converted into an equivalent state space realization, and this state space realization is combined with the external network equations of the offshore wind power flexible DC grid-connected system to eliminate algebraic variables and construct a closed-loop state space model describing all dynamic characteristics of the system. The state matrix of the closed-loop state-space model is decomposed into eigenvalues ​​to obtain all eigenvalues. From all eigenvalues, several eigenvalues ​​with the largest real parts that reflect the overall stability trend of the system are selected as the dominant poles of the system.

5. The method according to claim 1, characterized in that, S6 includes: A real part stability boundary threshold is preset, which is a negative number or zero; Extract the real part value of each pole in the dominant poles of the system one by one, and compare each real part value with the real part stability boundary threshold; When the real part values ​​of all dominant poles of the system are less than the real part stability boundary threshold, an evaluation result is generated to characterize the dynamic stability of the offshore wind power flexible DC grid-connected system under the current operating conditions. When the real part of at least one dominant pole of the system is greater than or equal to the real part stability boundary threshold, an assessment result is generated that characterizes the risk of instability of the offshore wind power flexible DC grid-connected system under the current operating conditions.

6. A stability evaluation system for offshore wind power flexible DC grid connection system based on multi-access point simulation, used to implement the method according to any one of claims 1-5, characterized in that, include: The model building unit is used to build an electromagnetic transient simulation model of an offshore wind power flexible DC grid-connected system. This electromagnetic transient simulation model integrates an equivalent model of an offshore wind farm aggregation, a detailed model of a flexible DC transmission system, and an equivalent model of a receiving-end AC grid. The disturbance injection unit is used to locate multiple physical connection points between the aggregated equivalent model of offshore wind farm and the detailed model of flexible DC transmission system in the electromagnetic transient simulation model, and to synchronously and in parallel inject power disturbance signals with preset amplitude and frequency at each physical connection point to stimulate the broadband dynamic response of the offshore wind power flexible DC grid-connected system across the entire access point range. The data acquisition unit is used to simultaneously acquire voltage timing response data and current timing response data at each physical connection point during the injection of power disturbance signals. The matrix construction unit is used to receive voltage timing response data and current timing response data, and to perform frequency domain transformation processing on the received data. Based on the transformed frequency domain data, a multi-port admittance matrix is ​​constructed to characterize the dynamic interactive coupling relationship between each physical connection point. The pole solving unit is used to receive the multi-port admittance matrix, perform frequency domain fitting on the multi-port admittance matrix, construct a continuous domain closed-loop state space model of the offshore wind power flexible DC grid-connected system, solve the eigenvalues ​​of the closed-loop state space model, and extract the dominant system poles that determine the dynamic characteristics of the system. The evaluation conclusion output unit is used to receive the dominant pole of the system, compare the real part of the dominant pole with the preset stability boundary threshold, and output the stability evaluation conclusion of the offshore wind power flexible DC grid connection system under the current operating conditions based on the comparison result.

7. The system according to claim 6, characterized in that, The power disturbance signal injected by the disturbance injection unit is specifically a set of sinusoidal power disturbance signals with interleaved frequencies applied simultaneously at each physical connection point. This set of interleaved sinusoidal power disturbance signals covers the wide frequency band to be analyzed, and the frequency difference between each sinusoidal signal is greater than a preset frequency resolution threshold.

8. The system according to claim 6, characterized in that, The matrix construction unit includes: The data preprocessing subunit is used to synchronize the voltage timing response data and current timing response data at each physical connection point, eliminate the time delay introduced by the difference in data acquisition channels, and obtain the synchronized time domain dataset. The frequency domain transformation subunit is used to perform a fast Fourier transform on the synchronized time domain dataset to obtain the frequency domain voltage response value and frequency domain current response value at each frequency point at each physical connection point. The matrix element solving sub-unit is used to construct the admittance matrix element corresponding to each frequency point in the multi-port admittance matrix by solving a system of linear equations based on the frequency domain voltage response value and frequency domain current response value at each frequency point.

9. The system according to claim 6, characterized in that, The pole solving unit includes: The fitting approximation sub-unit is used to approximate each matrix element in the multi-port admittance matrix by rational fractional combination using the vector matching method, so as to obtain a set of frequency domain response characteristic expressions that share the same pole set; The model construction subunit is used to convert the multi-port admittance matrix into an equivalent state-space realization based on the set of frequency domain response characteristic expressions, and to combine the state-space realization with the external network equations of the offshore wind power flexible DC grid-connected system to eliminate algebraic variables and construct a closed-loop state-space model describing all dynamic characteristics of the system. The feature extraction subunit is used to perform eigenvalue decomposition on the state matrix of the closed-loop state-space model to obtain all eigenvalues, and selects several eigenvalues ​​with the largest real parts that reflect the overall stability trend of the system as the dominant poles of the system.

10. The system according to claim 6, characterized in that, The evaluation conclusion output unit includes: The threshold preset subunit is used to set a real part stability boundary threshold, which is a negative number or zero. The comparison sub-unit is used to extract the real part value of each pole in the dominant poles of the system one by one, and compare each real part value with the real part stability boundary threshold. The conclusion generation sub-unit is used to generate an assessment result characterizing the dynamic stability of the offshore wind power flexible DC grid-connected system under the current operating conditions when the real part values ​​of all dominant poles of the system are less than the real part stability boundary threshold, based on the comparison results of the comparison sub-unit. When the real part value of at least one dominant pole of the system is greater than or equal to the real part stability boundary threshold, an assessment result characterizing the risk of instability of the offshore wind power flexible DC grid-connected system under the current operating conditions is generated.

Images