Grid-connected analysis method for time-varying impedance model of deep-sea wind power based on time-domain residual method

Abstract

This application discloses a grid-connected analysis method for time-varying impedance models of deep-sea wind power based on the time-domain residual method, belonging to the field of power system modeling and simulation technology. This method enables efficient and accurate analysis of grid-connected deep-sea wind power. The method includes: establishing a system of differential-algebraic equations containing high-capacity submarine cable impedance and multiple nonlinear elements, and constructing a recursive coefficient matrix; dividing a preset global time domain into multiple time-domain sub-intervals, and solving the recursive coefficient matrix in each time-domain sub-interval using a general analytical solution expressed by a power series explicit operator, generating a global time-domain explicit solution corresponding to the preset global time domain; performing a discrete Fourier transform on the global time-domain explicit solution to obtain time-frequency domain spectrum results; constructing a time-varying impedance model based on the differential-algebraic equations, and tuning the time-varying impedance model using the time-frequency domain spectrum results; and using the tuned time-varying impedance model to analyze the grid-connected control system of deep-sea wind power.

Description

Technical Field

[0001] This application relates to the field of power system modeling and simulation technology, and in particular to a grid-connected analysis method for a time-varying impedance model of deep-sea wind power based on the time-domain residual method. Background Technology

[0002] With the transformation of the global energy structure and the continuous growth in demand for clean energy, offshore wind power, as an important renewable energy source, is gradually becoming a research hotspot and development direction in the energy field. Offshore wind power is connected to the grid via high-capacity cables and power electronic converters. Due to the widespread application of power electronic converters, the system inertia is significantly reduced, while the grid's equivalent impedance changes rapidly and nonlinearly with operating conditions. Therefore, accurately grasping the time-varying impedance characteristics of the source and grid during offshore wind power grid connection is crucial for ensuring the stable operation of the power grid.

[0003] In related technologies, existing technical solutions for setting time-varying impedance models of source networks mainly include numerical methods, analytical methods, and machine learning methods. Numerical methods utilize dense micro-step integrals to solve the electromagnetic transient state equations containing time-varying impedances. For example, algorithms such as ODE23 and ODE45 integrated in MATLAB / Simulink are used. When simulating deep-sea grid-connected systems, the step size typically needs to be maintained at the microsecond level to avoid algebraic loops and achieve numerical convergence. Analytical methods establish s-domain impedance models through Laplace transformation of the time-domain state-space equations, effectively reproducing and analyzing the causes of post-fault events. Machine learning methods perform data analysis based on large amounts of recorded waveform data, fitting black-box impedance mappings. A typical process includes sampling, labeling, and training.

[0004] However, the applicant recognizes that existing technical solutions for tuning time-varying impedance models of source-grid systems have many defects and technical problems. Traditional numerical methods rely on dense micro-step numerical integration, which is a low-order approximation and faces computational bottlenecks when dealing with large-scale deep-sea power systems, making it difficult to efficiently handle highly complex problems. Analytical methods rely on linear operating points, lacking flexibility in characterization. Once the system state variables are linearized with small signals, they are "frozen," making it impossible to continuously and accurately track the equivalent impedance characteristics of the system during frequent power dispatching or fault-recovery processes. Machine learning methods are highly dependent on the scale of training data. If the sample coverage is insufficient or the operating conditions are extrapolated too far, the difference between the training impedance and the true impedance will be amplified by the model, easily leading to overfitting or misjudgment. Furthermore, the lack of decomposition of physical components makes it difficult to provide interpretable basis for engineering parameter tuning. Therefore, there is an urgent need to develop an efficient, accurate, and adaptable method for tuning time-varying impedance models of source-grid systems to enable efficient and accurate analysis of deep-sea wind power grid integration. Summary of the Invention

[0005] In view of this, this application provides a grid-connected analysis method for time-varying impedance models of deep-sea wind power based on the time-domain residual method. The main purpose is to solve the problems of computational bottlenecks in the current processing of large-scale deep-sea power systems, difficulty in efficiently dealing with high complexity, easy overfitting or misjudgment, lack of decomposition of physical components, and difficulty in providing interpretable basis for engineering parameter tuning.

[0006] According to the first aspect of this application, a grid-connected analysis method for time-varying impedance models of deep-sea wind power based on the time-domain residual method is provided, the method comprising:

[0007] A system of differential-algebraic equations containing high-capacity submarine cable impedance and multiple nonlinear elements is established. A recursive coefficient matrix is ​​constructed by utilizing the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system. The recursive coefficient matrix is ​​used to describe the state change law of the deep-sea wind power grid-connected control system at multiple times.

[0008] The preset global time domain is divided into multiple time domain sub-intervals, and the recursive coefficient matrix is ​​solved in each time domain sub-interval by referring to the general analytical solution represented by the power series explicit operator, thereby generating the global time domain explicit solution corresponding to the preset global time domain.

[0009] The time-domain explicit solution is subjected to a discrete Fourier transform to obtain the time-frequency domain spectral results.

[0010] A time-varying impedance model is constructed based on the differential algebraic equation, and the time-varying impedance model is tuned using the time-frequency domain spectral results. The tuned time-varying impedance model is then used to analyze the deep-sea wind power grid-connected control system.

[0011] Using the above technical solution, this application provides a grid-connected analysis method for a time-varying impedance model of deep-sea wind power based on the time-domain residual method. This application establishes a system of differential-algebraic equations including the impedance of a large-capacity submarine cable and multiple nonlinear elements. It then utilizes the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system to construct a recursive coefficient matrix. The preset global time domain is divided into multiple time-domain sub-intervals. Referring to a general analytical solution expressed using a power series explicit operator, the recursive coefficient matrix is ​​solved in each time-domain sub-interval to generate a global time-domain explicit solution corresponding to the preset global time domain. The global time-domain explicit solution is then discretized. Fourier transform is used to obtain the time-frequency domain spectrum results. A time-varying impedance model is constructed based on the differential algebraic equations, and the time-varying impedance model is tuned using the time-frequency domain spectrum results. The tuned time-varying impedance model is then used to analyze the grid-connected control system of deep-sea wind power. By solving the full-time domain multi-segment sub-interval explicit solution and calculating the time-frequency domain spectrum results, the microstep integration is significantly reduced, the consumption of computational resources is reduced, and the computational efficiency is improved. Moreover, the explicit time-domain solution can be refreshed online according to the operating conditions, quickly capturing the stability changes of deep-sea grid connection, improving the characterization accuracy and adaptability, and has high applicability in the context of large-scale development of deep-sea wind power.

[0012] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0013] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0014] Figure 1 This paper illustrates a schematic diagram of a grid-connected analysis method for a time-varying impedance model of deep-sea wind power based on the time-domain residual method, provided in an embodiment of this application.

[0015] Figure 2 A schematic diagram of the circuit system of a deep-sea wind power grid-connected control system provided in an embodiment of this application is shown;

[0016] Figure 3 This paper presents a schematic diagram of another grid-connected analysis method for time-varying impedance models of deep-sea wind power based on the time-domain residual method provided in an embodiment of this application. Detailed Implementation

[0017] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0018] This application provides a grid-connected analysis method for a time-varying impedance model of deep-sea wind power based on the time-domain residual method, such as... Figure 1 As shown, the method includes:

[0019] S10. Establish a set of differential-algebraic equations that include the impedance of large-capacity submarine cables and multiple nonlinear elements, and construct a recursive coefficient matrix by utilizing the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system.

[0020] The embodiments of this application can be applied to a power grid analysis system. This system is equipped with a server, and the server's computing power can be used to implement the technical solution of this application. Specifically, the power grid analysis system can analyze the grid-connected control system for deep-sea wind power, and the circuit system of the deep-sea wind power grid-connected control system is as follows: Figure 2 As shown, on the left is a DC voltage source Vdc obtained from wind power generation via rectification (DER module). After passing through a filter circuit containing inductor Lf and capacitor Cf, it is connected to the relevant grid components. The circuit system includes coordinate transformation modules, such as "abc / dq0" and "dq0 / abc", used to convert electrical quantities between different coordinate systems for better control analysis. Furthermore, the circuit system contains current and voltage measurement and feedback paths, such as iabc and vabc signals, which are transformed before participating in control calculations. In addition, the control loop of the circuit system includes multiple transfer function modules Gi(s) and Gv(s) for regulating the current and voltage of the circuit system, a "Power Cal" module for power calculation, and a "Primary Regulator" as the main regulator, which can adjust the circuit system based on reference values ​​such as vref_dq0. The circuit system is also connected to point-to-point control (PCC) and grid impedance Zg, forming a complete grid-connected control closed-loop system to achieve efficient and stable grid connection of wind power generation.

[0021] To analyze the grid state during deep-sea wind power grid connection, in this embodiment, the grid analysis system establishes a system of differential-algebraic equations including the impedance of a large-capacity submarine cable and multiple nonlinear components. The large-capacity submarine cable impedance is a quantitative representation of the cable's resistance to current in the deep-sea wind power grid connection control system, influenced by factors such as cable length, material, cross-sectional area, and current frequency. The multiple nonlinear components encompass parts of the deep-sea wind power grid connection control system that exhibit nonlinear characteristics, such as power electronic devices; the output of these components is not directly proportional to the input. By deeply studying the physical structure and working principle of the deep-sea wind power grid connection control system, collecting parameters of each component within the system, and based on circuit theory, electromagnetic theory, and other knowledge, the grid analysis system can construct a system of differential-algebraic equations that accurately describes the dynamic characteristics of the deep-sea wind power grid connection control system.

[0022] Simultaneously, the power grid analysis system also utilizes the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system, such as the correlation between voltage, current, and power variables across different components and at different times, to construct a recursive coefficient matrix. This recursive coefficient matrix describes the state change patterns of the deep-sea wind power grid-connected control system at multiple times. It links the current state of the control system to the state at the next or subsequent times, providing a foundation for subsequent analysis of the state evolution of the control system.

[0023] Thus, by constructing a system of differential-algebraic equations and a recursive coefficient matrix, a precise mathematical model is provided for in-depth analysis of the grid-connected control system of deep-sea wind power. This model accurately characterizes the dynamic characteristics of the control system at different times, laying a solid foundation for subsequent solutions and analyses. For example, in a certain deep-sea wind farm project, the submarine cable length reaches tens of kilometers, and various power electronic devices of different specifications are used. By establishing a system of differential-algebraic equations that includes the impedance of large-capacity submarine cables and multiple nonlinear elements, and constructing a recursive coefficient matrix, the changes in state variables such as voltage and current in the wind power grid-connected system at different times can be accurately described, providing accurate data support for subsequent analysis of system stability.

[0024] Step S10, which involves establishing a system of differential-algebraic equations that includes the impedance of a large-capacity submarine cable and multiple nonlinear elements, includes the following steps:

[0025] S11. Identify multiple nonlinear elements, including a PLL phase-locked loop, a droop control loop, and a filtering loop.

[0026] When constructing the differential-algebraic equations for a grid-connected control system for deep-sea wind power, the power grid analysis system first identifies several nonlinear components. These components include a PLL (phase-locked loop), droop control, and filtering. The PLL enables the system to accurately track the phase and frequency of the input signal, ensuring synchronization with the grid during grid connection. The droop control regulates active and reactive power to control system voltage and frequency, maintaining stable system operation. The filtering component removes harmonics and noise from the system, improving power quality.

[0027] Identifying these nonlinear elements is fundamental to constructing an accurate system of differential-algebraic equations. This helps to comprehensively describe the dynamic characteristics of the deep-sea wind power grid-connected control system, provides key information for the subsequent construction of an accurate system of differential-algebraic equations, and enables the equations to more comprehensively and accurately reflect the complex dynamic behavior of the deep-sea wind power grid-connected control system, thus laying a solid foundation for the system's analysis and control.

[0028] S12. Referring to the local and global relationships of the inverter in the dq coordinate system, construct the equations for the PLL phase-locked loop.

[0029] In power systems, the dq coordinate system is a commonly used coordinate system that converts three-phase AC quantities into DC quantities, facilitating analysis and control. The power grid analysis system constructs the PLL (Phase-Locked Loop) equations by referring to the local and global relationships between the inverter and the power grid in the dq coordinate system. The PLL equations are shown in Equation 1 below:

[0030] Formula 1:

[0031] in, It is the phase difference of the phase-locked loop, ω dq It is the local angular velocity of the inverter in the dq coordinate system, ω DQ This is the global reference angular velocity of the power grid in the dq coordinate system, which is also the desired angular velocity value of the system. By constructing the above equations, the dynamic behavior of the phase-locked loop (PLL) in the dq coordinate system can be described, providing a mathematical model for the phase synchronization control of the system. This helps to achieve precise synchronization between the system and the power grid, improving the stability and reliability of the deep-sea wind power grid-connected control system. For example, when connecting deep-sea wind power to the grid, the frequency and phase of the power grid may fluctuate due to factors such as sea waves. The PLL equations can be used to adjust the phase of the system in real time, ensuring that the wind power system remains synchronized with the power grid and reducing the occurrence of power oscillations and loss of synchronization.

[0032] S13. Construct the equation system for the droop control link.

[0033] The equation set for the droop control element constructed by the power grid analysis system can be found in Equation 2 below:

[0034] Formula 2:

[0035] Among them, P m Q is active power, representing the actual active power output by the system; m It is reactive power, representing the actual reactive power output of the system; G P and G Q It is the droop control gain, used to adjust the degree of response of active and reactive power to changes in system parameters; T P and T Q It is the time constant, which determines the response speed of droop control; P is the given reference active power, and Q is the given reference reactive power.

[0036] By constructing a set of equations for the droop control mechanism, we can describe how the droop control mechanism adjusts according to the actual power output of the system to maintain voltage and frequency stability. This provides an accurate mathematical model for power control, contributing to stable system operation and rational power allocation, and improving the performance of deep-sea wind power systems. For example, in deep-sea wind farms, when wind speed changes cause power fluctuations, the droop control mechanism can adjust the active and reactive power outputs in real time according to the equations to maintain the system voltage and frequency within allowable ranges, ensuring stable system operation.

[0037] S14. Based on the impedance of high-capacity submarine cables, construct a set of equations for the filtering stage.

[0038] The set of equations for the filtering stage in the power grid analysis system can be found in Equation 3 below:

[0039] Formula 3:

[0040] Among them, v d and v q It is the dq component of the inverter output bus voltage, i d and i q It is the dq component of the inverter output current through the filter inductor, i bd and i bq It is the dq component of the output grid-connected current through the grid-connected impedance, e d and e q It is the dq component of the controller output voltage, w dq It is the local angular velocity of the inverter, C f It is the filter capacitor, L. f It is the filter inductor of the filter.

[0041] The impedance of high-capacity submarine cables can affect the electrical characteristics of a system. Therefore, in this embodiment, constructing a set of equations for the filtering stage based on the impedance of the high-capacity submarine cable can more accurately describe the dynamic behavior of the filtering stage in a deep-sea wind power grid-connected system. This allows the subsequently constructed model to be closer to the actual system, improving the accuracy of predicting the system's harmonic and noise filtering effects and ensuring the power quality of deep-sea wind power grid connection. For example, during deep-sea wind power transmission, high-capacity submarine cables generate significant impedance, leading to current and voltage distortion. By constructing a set of equations for the filtering stage based on the submarine cable impedance, filter parameters can be better designed, harmonics can be effectively filtered out, and power quality can be improved.

[0042] S15. Combine the equations of the PLL phase-locked loop, the droop control loop, and the filter loop to obtain a set of differential algebraic equations.

[0043] After constructing the equations through the above process, the power grid analysis system combines the equations for the PLL phase-locked loop, the droop control loop, and the filtering loop to obtain a set of differential-algebraic equations. This set of differential-algebraic equations integrates multiple key components and can comprehensively describe the dynamic characteristics of the deep-sea wind power grid-connected system. By combining these equations, a unified mathematical model can be established to analyze the system's behavior under different operating conditions, providing a complete mathematical tool for the analysis and control of deep-sea wind power grid-connected systems. This allows for more accurate prediction of the system's response and provides strong support for the system's optimized design and stable operation. For example, during the design and commissioning phases of the deep-sea wind power grid-connected system, this set of differential-algebraic equations can be used for simulation analysis to predict the system's performance under different wind speeds, sea states, and other conditions, promptly identify potential problems, and optimize the system to improve its reliability and stability.

[0044] In step S10, which involves constructing a recursive coefficient matrix using the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system, the following steps are included:

[0045] S16. Based on the relationship between multiple state variables in the deep-sea wind power grid-connected control system, the full state quantity mapping constraint matrix is ​​derived.

[0046] In the analysis of deep-sea wind power grid-connected control systems, the grid analysis system derives the full-state mapping constraint matrix based on the relationships between multiple state variables within the control system. The deep-sea wind power grid-connected control system contains numerous state variables, such as the wind turbine's rotational speed, output voltage, and current, as well as the grid's voltage and frequency. These state variables have complex physical relationships; for example, the wind turbine's rotational speed affects the frequency and amplitude of its output voltage and current, while the grid's voltage and frequency provide feedback to the wind turbine's operation. The full-state mapping constraint matrix reflects the linear constraint relationships between multiple state variables. By deeply analyzing the intrinsic connections between the state variables in the deep-sea wind power grid-connected control system, and using methods such as linear algebra, these relationships are expressed in matrix form. Specifically, the grid analysis system determines the linear combination relationships between the state variables based on the physical laws and operating characteristics of the deep-sea wind power grid-connected control system, thereby constructing the full-state mapping constraint matrix. See Formula 4 below for the full-state mapping constraint matrix.

[0047] Formula 4:

[0048] Where A represents the full-state mapping constraint matrix, used to reflect the linear and nonlinear constraint relationships between multiple state variables; X represents the vector composed of state variables. Equation 4 expresses the relationship between the matrix and the state variable vector, used to describe the constraints. Accurately deriving the full-state mapping constraint matrix can clearly describe the intrinsic connections between various state variables in the deep-sea wind power grid-connected control system, providing a foundation for the subsequent construction of the recursive coefficient matrix, helping to more accurately analyze the dynamic characteristics of the system and improve the accuracy of predicting system behavior. For example, in a certain deep-sea wind farm project, the connection between the wind turbine and the grid is complex, with multiple state variables interacting with each other. By deriving the full-state mapping constraint matrix, the linear constraint relationships between various state variables can be clarified, such as how wind speed changes affect the wind turbine speed, and thus affect the output voltage and current, and how these changes are fed back to the grid side, providing key information for subsequent system analysis and control.

[0049] S17. Construct the recursive coefficient matrix by referring to the full state quantity mapping constraint matrix.

[0050] The power grid analysis system constructs a recursive coefficient matrix by referring to the full state variable mapping constraint matrix. The recursive coefficient matrix is ​​used to describe the recursive relationship between state variables at different times. The specific recursive coefficient matrix can be found in Formula 5 below:

[0051] Formula 5:

[0052] Among them, a ijThe elements in the recursive coefficient matrix are represented by i, where i represents the row and j represents the column; x i [k] represents the k-th order coefficient of the i-th state variable, where k is a natural number; x i [k+1] represents the (k+1)th order coefficient of the i-th state variable, calculated using the recursive coefficient matrix and the current state variable value. When constructing the recursive coefficient matrix, the power grid analysis system determines the values ​​of each element based on the linear relationships between state variables revealed by the full-state mapping constraint matrix, combined with the system's dynamic characteristics. These coefficients reflect the changing patterns of state variables from the current time to the next time step. Using the recursive coefficient matrix and the current state variable value, the state variable value for the next time step can be calculated. Thus, constructing the recursive coefficient matrix establishes the connection between state variables at different times, enabling recursive calculations of the system's dynamic evolution. This helps track system state changes in real time, providing strong support for real-time system control and optimization, and overcoming the computational bottleneck caused by microstep integration. For example, in the operation of a deep-sea wind power grid-connected system, by constructing a recursive coefficient matrix, the values ​​of these state variables, such as the wind turbine's rotational speed and output voltage, can be recursively calculated for the next moment. This allows for early prediction of the system's operating state, timely adjustment of control strategies, and ensuring stable system operation. When wind speed changes suddenly, the recursive coefficient matrix can be used to quickly predict changes in the wind turbine's output power, thereby adjusting the grid's access power and avoiding impacts on the grid.

[0053] S20. Divide the preset global time domain into multiple time domain sub-intervals, and with reference to the general analytical solution represented by the power series explicit operator, solve the recursive coefficient matrix in each time domain sub-interval to generate the global time domain explicit solution corresponding to the preset global time domain.

[0054] In this embodiment, the power grid analysis system divides a preset global time domain into multiple time domain sub-intervals. The preset global time domain refers to the entire time range set for analyzing the deep-sea wind power grid-connected control system, such as the entire process from system startup to a period of stable operation. Dividing the time domain into sub-intervals allows for a more detailed analysis of the state changes of the deep-sea wind power grid-connected control system over different time periods. The number and length of the sub-intervals can be determined based on factors such as the characteristics of the deep-sea wind power grid-connected control system, research needs, and computational accuracy requirements.

[0055] Then, the power grid analysis system refers to the general analytical solution expressed using power series explicit operators to solve the recursive coefficient matrix in each time domain sub-interval. The power series explicit operator is a mathematical tool that can transform complex problems into power series forms for solution. The general analytical solution is a form of general solution applicable to this type of problem. By applying the power series explicit operator to the recursive coefficient matrix, combined with constraints such as the boundary conditions of the sub-intervals, the power grid analysis system will progressively calculate the state changes of the system within each sub-interval, ultimately generating a global time domain explicit solution corresponding to the preset global time domain.

[0056] By dividing the global time domain into multiple sub-intervals for solution, the enormous computational burden of directly solving complex problems across the entire time domain is avoided. This significantly reduces the number of microstep integrations, thereby lowering computational resource consumption and improving computational efficiency. For example, when analyzing the operation of a deep-sea wind power grid-connected control system within a day, the day is divided into multiple hourly segments as time domain sub-intervals. Within each sub-interval, the recursive coefficient matrix is ​​solved using power series explicit operators, allowing for rapid acquisition of the system's state changes within that sub-interval. This generates a global time domain explicit solution for the entire day, greatly shortening computation time and improving analytical efficiency compared to traditional methods.

[0057] In step S20, which involves solving the recursive coefficient matrix in each time domain sub-interval using the general analytical solution expressed by the power series explicit operator, and generating the global time domain explicit solution corresponding to the preset global time domain, the following steps are included:

[0058] S21. In each time domain subinterval, the general analytical solution of the time domain subinterval is represented by a power series explicit operator.

[0059] The power grid analysis system uses power series explicit operators to represent the general analytical solution for each time-domain sub-interval. A power series explicit operator is a mathematical expression that presents complex time-domain solutions in the form of power series, facilitating subsequent calculations and analysis. Specifically, the general analytical solution for each time-domain sub-interval can be represented using power series explicit operators according to the following formula 6:

[0060] Formula 6: x(t) = x(t) k )+Σφ (k) (tt k ) k

[0061] Where x(t) represents the state vector of the deep-sea wind power grid-connected control system at time t, which encompasses various key state information of the system at that time, such as voltage, current, and power; x(t k ) represents the current time domain sub-interval [t] k ,t k+1The initial state vector, i.e., the state of the system at the beginning of this sub-interval; φ (k) represents a set of explicitly expanded coefficients constructed based on the dynamic characteristics of the deep-sea wind power grid-connected control system within the current time domain sub-interval. These coefficients are recursively derived from the dynamic equations and boundary conditions of the deep-sea wind power grid-connected control system, reflecting the dynamic characteristics of the system within this sub-interval; t represents the current time, with values ​​ranging from [t...]. k ,t k+1 [Inner; t] k This indicates the start time of the current time domain sub-interval.

[0062] Using Formula 6 above, the power grid analysis system can obtain a general solution expression in each time domain sub-interval, transforming the complex time-domain problem into a relatively simple power series form, simplifying the calculation process, and preserving the dynamic characteristic information of the system. This provides an effective way to accurately solve the system in each time domain sub-interval. For example, due to the complexity of sea conditions, the system state changes in a complex manner over time. This method can more clearly describe the changes in the state of the deep-sea wind power grid-connected control system in each small time domain sub-interval.

[0063] S22. The Pade approximation algorithm is used to extend the convergence domain of the general analytical solution for each time domain subinterval to obtain an expression, and the explicit time domain power series solution for each time domain subinterval is determined based on the expression.

[0064] The power grid analysis system employs the Pade approximation algorithm to process the general analytical solution for each time-domain sub-interval. The Pade approximation algorithm is a method that transforms a power series into a rational function, expanding the approximate convergence region and improving approximation accuracy. This algorithm yields an expression, and based on this expression, the explicit solution of the time-domain power series for each time-domain sub-interval is determined. This expression is detailed in Formula 7 below.

[0065] Formula 7:

[0066] Where x(t) represents the state vector of the deep-sea wind power grid-connected control system at time t; P M It is the numerator polynomial, used to extend the approximate region of convergence, and its coefficients p m Determined based on the Pade approximation algorithm; Q N It is the denominator polynomial, used to extend the approximate region of convergence; p m It is the molecule polynomial P M The coefficient of q; n It is the denominator polynomial Q N The coefficient; t represents the current time, and its value ranges within the current time domain subinterval [t]. k ,t k+1 [Inner; t] kThis represents the reference time point, which serves as the benchmark for time sampling in the Pade approximation algorithm.

[0067] After processing with the Pade approximation algorithm, the resulting explicit time-domain power series solution can more accurately describe the state changes of the system within that time-domain sub-interval, effectively improving the convergence of the power series and increasing the approximation accuracy of the system state solution. This makes the obtained explicit time-domain power series solution closer to the actual system state, providing more reliable data support for subsequent overall system analysis. For example, in deep-sea wind power grid-connected control systems, due to various disturbances and uncertainties, processing with the Pade approximation algorithm can more accurately obtain the system state solution in each time-domain sub-interval, reducing errors.

[0068] S23. By defining an error loss function, estimate the adaptive order and step size so that the explicit solution of the time-domain power series of each time-domain sub-interval remains convergent within a preset error threshold.

[0069] In this embodiment, the power grid analysis system estimates the adaptive order and step size by defining an error loss function. The error loss function measures the degree of error between the explicit solution of the time-domain power series and the actual system state. Adjustments to the adaptive order and step size ensure that the explicit solution of the time-domain power series within each time-domain sub-interval remains convergent within a preset error threshold. In deep-sea wind power grid-connected control systems, due to the complex and variable system state, different time-domain sub-intervals may require different orders and step sizes to ensure solution accuracy. This embodiment continuously adjusts the order and step size to find the most suitable parameter combination for the current time-domain sub-interval, enabling the solution to better approximate the actual system state.

[0070] In this way, by adaptively adjusting the order and step size, the calculation process can be optimized and the calculation efficiency improved, while ensuring solution convergence, based on the actual dynamic characteristics of the system. This also ensures that the accuracy of the solution meets requirements, making the analysis results more reliable. For example, in the operation of a deep-sea wind power grid-connected system, when the system is under different operating conditions, adaptive parameter adjustment can more accurately obtain the system state solution and promptly detect potential anomalies.

[0071] S24. Concatenate the explicit solutions of multiple time-domain power series from multiple time-domain subintervals to obtain the global explicit solution in the time domain.

[0072] The power grid analysis system will concatenate multiple explicit time-domain power series solutions from multiple time-domain sub-intervals to obtain a global explicit time-domain solution. Specifically, the concatenation of multiple explicit time-domain power series solutions can be achieved using the following formula 8:

[0073]

[0074] Where X(t) represents the state vector of the deep-sea wind power grid-connected control system at time t in the global time domain; U represents the splicing operation, which splices multiple explicit solutions of time-domain power series to form an explicit solution in the global time domain; k represents the index of the time-domain sub-interval, with values ​​from 1 to N, used to identify each time-domain sub-interval; x k (t) represents the time domain subinterval [t] k ,t k+1 The local approximate solution; the range of values ​​for t is [t]. k ,t k+1 [Inner; t] k Let t represent the starting time of the k-th time-domain sub-interval. k+1 This represents the end time of the k-th time domain sub-interval.

[0075] Through splicing operations, the power grid analysis system can integrate the solutions from various time-domain sub-intervals to form a complete description of the deep-sea wind power grid-connected control system across the entire global time domain. This comprehensively reflects the dynamic changes of the control system throughout the entire analysis time domain, providing complete solution information for the overall system analysis and evaluation. It helps to gain a deeper understanding of the system's operating characteristics and stability, providing a strong basis for system optimization and control. For example, in the long-term operation analysis of the deep-sea wind power grid-connected system, the explicit global time-domain solution obtained through splicing clearly shows the trend of system state changes in different time periods, providing support for formulating reasonable operation and maintenance strategies.

[0076] S30. Perform a discrete Fourier transform on the global time-domain explicit solution to obtain the time-frequency domain spectrum result.

[0077] In this embodiment, the power grid analysis system performs a Discrete Fourier Transform (DFT) on the global time-domain explicit solution. The DFT is a mathematical method that converts a time-domain signal into a frequency-domain signal, decomposing a discrete data sequence in the time domain into a superposition of sine and cosine components of different frequencies. Through the DFT, the power grid analysis system can transform the global time-domain explicit solution from the time domain to the frequency domain, obtaining a time-frequency domain spectrum result. The time-frequency domain spectrum result clearly demonstrates the energy distribution of different frequency components in the deep-sea wind power grid-connected control system, helping to analyze the dynamic characteristics of the deep-sea wind power grid-connected control system at different frequencies, such as the presence of harmonics, oscillations, and other problems.

[0078] Obtaining time-frequency domain spectra through Discrete Fourier Transform (DFT) allows for in-depth analysis of the characteristics of deep-sea wind power grid-connected control systems from a frequency domain perspective. This reveals issues difficult to detect in time-domain analysis, providing crucial information for comprehensively evaluating the stability and performance of these systems. For instance, when analyzing the power quality of a deep-sea wind power grid-connected control system, performing a DFT on the global time-domain explicit solution yields time-frequency domain spectra. These spectra clearly show the presence of harmonics at specific frequencies within the control system, as well as the energy levels of each frequency component. This allows for assessment of whether the system's power quality meets standard requirements, providing guidance for implementing appropriate improvement measures.

[0079] In step S30, which involves performing a discrete Fourier transform on the global time-domain explicit solution to obtain the time-frequency domain spectrum result, the following steps are included:

[0080] S31. Determine the specified step size and use the specified step size as the time window.

[0081] In the time-domain spectral analysis of deep-sea wind power grid-connected systems, the grid analysis system first needs to determine a specified step size, which is then used as the time window. The specified step size is a pre-set time interval based on the system's analytical requirements and characteristics; it determines the time resolution of subsequent discrete Fourier transforms. For example, when analyzing the dynamic characteristics of a deep-sea wind power grid-connected control system, if the focus is on rapidly changing signals, a smaller step size is needed to capture signal changes more precisely; if the focus is on slower trends, the step size can be appropriately increased. Using the specified step size as the time window means that subsequent analyses will be performed in units of this time interval.

[0082] Specifically, the power grid analysis system pre-derives the time-domain power-order numerically explicit solution to the spectrum mapping module, embeds the discrete Fourier transform into the time-domain power-order numerically explicit solution, and defines the time window by setting the step size. In this way, by determining an appropriate specified step size as the time window, a balance can be struck between time resolution and computational load, providing a suitable time scale for the subsequent discrete Fourier transform. This allows the analysis to capture the important dynamic characteristics of the system without compromising efficiency due to excessive computational load. For example, in a deep-sea offshore wind power grid-connected system, when analyzing the impact of sudden wind speed changes on the system, setting a smaller step size allows for more accurate observation of the system's response at the instant of wind speed change, providing a basis for timely adjustment of control strategies.

[0083] S32. Within each time window, perform a discrete Fourier transform on the time-domain coefficients of the global time-domain explicit solution to convert the time-domain coefficients into frequency-domain coefficients, and with reference frequency-domain coefficients, generate time-frequency domain spectrum results.

[0084] Within each defined time window, the power grid analysis system performs a Discrete Fourier Transform (DFT) on the time-domain coefficients of the global explicit time-domain solution. The DFT is an important method for converting time-domain signals into frequency-domain signals; it decomposes discrete data sequences in the time domain into a superposition of sine and cosine components at different frequencies. Through the DFT, the power grid analysis system converts the time-domain coefficients into frequency-domain coefficients, which reflect the energy distribution of the signal at different frequencies. Referring to these frequency-domain coefficients, a time-frequency domain spectrum result can be generated, specifically the amplitude-spectrum, which visually displays the signal amplitude at various frequencies. This lays the foundation for subsequent rapid frequency-domain analysis of grid-connected stability based on the updated time-frequency spectrum result.

[0085] By converting time-domain information into frequency-domain information through Discrete Fourier Transform (DFT), the power grid analysis system can conduct in-depth analysis of the characteristics of deep-sea wind power grid-connected systems from a frequency domain perspective. This allows for the discovery of problems that are difficult to detect in time-domain analysis, such as the presence of harmonics or oscillations at specific frequencies. This provides crucial information for a comprehensive assessment of the system's stability and performance and facilitates rapid frequency-domain analysis, improving analysis efficiency. For example, in deep-sea wind power grid-connected systems, the generated time-frequency domain spectrum results clearly show whether harmonic components with frequencies close to the grid exist. If present, these components may enhance the interaction between the system and the grid, affecting grid stability. In such cases, appropriate measures can be taken to suppress them.

[0086] S40. Construct a time-varying impedance model based on differential algebraic equations, and tune the time-varying impedance model using time-frequency domain spectral results. Analyze the grid-connected control system of deep-sea wind power using the tuned time-varying impedance model.

[0087] In this embodiment, the power grid analysis system constructs a time-varying impedance model based on differential-algebraic equations. This time-varying impedance model considers the time-varying characteristics of the parameters of the deep-sea wind power grid-connected control system, and can more accurately reflect the impedance characteristics of the control system at different times. Simultaneously, based on the previously established differential-algebraic equations and the actual operating conditions of the deep-sea wind power grid-connected control system, the power grid analysis system determines the time-varying variation patterns of each parameter in the model, thereby constructing the time-varying impedance model.

[0088] After obtaining the time-varying impedance model, the power grid analysis system uses time-frequency domain spectral results to tune the model. Tuning refers to adjusting and optimizing the model parameters based on actual measurement data or analysis results, making the model more accurately reflect the characteristics of the actual deep-sea wind power grid-connected control system. Specifically, the power grid analysis system can compare and analyze the frequency components and energy distribution information in the time-frequency domain spectral results with the time-varying impedance model, adjusting the model parameters to make the model output more consistent with the characteristics of the actual system in the frequency domain. In this way, the tuned time-varying impedance model can be used to analyze the deep-sea wind power grid-connected control system, enabling a more accurate evaluation of the stability, power transmission capacity, and other performance indicators of the deep-sea wind power grid-connected control system under different operating conditions.

[0089] By constructing and tuning a time-varying impedance model, the embodiments of this application enable the constructed time-varying impedance model to better adapt to changes in system parameters over time, improving the model's characterization accuracy and adaptability. Furthermore, the explicit time-domain solution can be updated online according to operating conditions, quickly capturing changes in grid-connected stability in deep-sea areas. This allows for timely tuning and analysis of the model using new data, demonstrating high applicability in the context of large-scale development of deep-sea wind power. For example, when the output power of a deep-sea wind farm fluctuates due to changes in wind speed, the system parameters will also change accordingly. By constructing a time-varying impedance model based on differential-algebraic equations and tuning the model using time-frequency domain spectral results, the model can accurately reflect the impedance characteristics of the system under different power output conditions. Analyzing the system using the tuned model can promptly identify potential stability issues during power fluctuations, providing strong support for implementing appropriate control strategies to ensure stable system operation.

[0090] In step S40, which involves tuning the time-varying impedance model using time-frequency domain spectral results, the following steps are included:

[0091] S41. Construct a simulation environment to simulate the grid connection operation of deep-sea wind power in the simulation environment, and generate characteristic frequency components in the simulation process by referring to the time-frequency domain spectrum results.

[0092] In the process of tuning the time-varying impedance model of a deep-sea wind power grid-connected system, the power grid analysis system first constructs a simulation environment. Specifically, this simulation environment can be constructed based on the benchmark model of a single-unit infinite-power grid-connected inverter. The single-unit infinite-power grid-connected inverter benchmark model is a commonly used power system analysis model. It assumes that the system consists of a generator and an infinite-power grid, which can simplify the analysis of complex power systems and provide a basic framework for studying deep-sea wind power grid-connected systems.

[0093] In the constructed simulation environment, the power grid analysis system sets up a step change in the grid-connected inductance for the droop-controlled inverter. The droop-controlled inverter plays a role in regulating power and stabilizing voltage and frequency in the deep-sea wind power grid-connected system, and the step change in the grid-connected inductance simulates the parameter abrupt changes that may occur in the actual system.

[0094] By referring to the time-frequency domain spectrum results, the power grid analysis system can solve for the time-domain waveforms of transient frequency and voltage response under operating conditions, obtain the time-domain analytical solution, and acquire the dynamic response data of the deep-sea wind power grid-connected control system in a simulation environment. The time-frequency domain spectrum results contain important information about the system in the frequency domain. By referring to it to solve for the time-domain waveform, the dynamic response data of the system under specific operating conditions can be obtained more accurately.

[0095] Simultaneously, within the hardware loop of the simulation environment, the power grid analysis system triggers a step change in the grid-connected inductor reactance and, referring to the time-domain analytical solution results, acquires the time-domain waveform of the deep-sea wind power grid-connected control system in the hardware loop in real time. The hardware loop more realistically simulates the operation of the actual system, and the real-time acquired time-domain waveform reflects the dynamic characteristics of the system in the actual hardware environment. By performing time-domain spectral analysis on the dynamic response data and time-domain waveforms, the power grid analysis system obtains characteristic frequency components. These characteristic frequency components are important manifestations of the system's dynamic characteristics and can help understand the system's response at different frequencies.

[0096] For example, in practical applications, a single-unit infinite bus grid-connected system can be selected as a benchmark experimental platform to conduct relevant tests. The specific process is as follows: First, based on the single-unit infinite bus grid-connected inverter benchmark model, and using MATLAB / Simulink as a simulation tool as a reference standard, the transient frequency and voltage response time-domain waveforms of the droop control inverter under the condition of a step change in the grid-connected inductance (from 0.10 per unit to 0.05 per unit) are accurately solved. This allows the simulation to obtain the dynamic response characteristics of the system under the ideal model. Next, to further verify the performance in a real hardware environment, the same set of control parameters used previously is downloaded to the OPAL-RT hardware-in-the-loop test platform (OP5600). In this hardware loop environment, a step change in the grid-connected inductor reactance is triggered again, and the test platform is used to collect the system's time-domain waveforms in real time. This obtains the dynamic response data of the system under near-real hardware operating conditions, thereby comprehensively evaluating the system performance.

[0097] S42. The model parameters of the time-varying impedance model are tuned by analyzing the characteristic frequency components.

[0098] In this embodiment, the power grid analysis system tunes the model parameters of the time-varying impedance model by analyzing the obtained characteristic frequency components. The time-varying impedance model describes the change in impedance characteristics between the source and grid in a deep-sea wind power grid-connected system over time. The accuracy of its model parameters directly affects the model's accuracy in describing system characteristics. Characteristic frequency components reflect the dynamic behavior of the system at different frequencies. By analyzing these components, the variation law of the system's impedance characteristics at different frequencies can be understood. Based on the analysis results, the parameters of the time-varying impedance model are adjusted so that the model can more accurately reflect the actual characteristics of the system. For example, if the characteristic frequency components show a large impedance change in the system within a certain frequency range, the parameters in the model related to that frequency range can be adjusted accordingly.

[0099] Tuning the time-varying impedance model using characteristic frequency components improves its accuracy and allows it to better adapt to dynamic system changes. This provides a more accurate basis for converter grid-connected stability analysis and optimization, facilitating the timely detection of potential stability issues and enabling corresponding improvements. This demonstrates the high efficiency of this application in real-time analysis of practical deep-sea grid-connected devices, providing strong support for the stable operation of deep-sea wind power grid-connected systems. Furthermore, compared to traditional methods, the proposed method for constructing the time-varying impedance model significantly reduces reliance on large-scale training data, optimizes data acquisition and processing, and lowers computational costs. Moreover, it adapts to dynamic changes in the context of high-penetration renewable energy, enabling real-time system adjustment and optimization to ensure system stability. It is suitable for real-time stability assessment of deep-sea wind power, distribution network inverters, and multi-bus systems, and can seamlessly integrate with OPAL-RT and other technologies.

[0100] In summary, the technical solution of this application is summarized as follows: See Figure 3 The input on the left is a time-domain waveform, which is processed to obtain a frequency-domain spectrum. The frequency-domain spectrum interacts with the time-domain waveform at the multiplication point. Combined with root / grid-type control, they are input to the impedance modeling module. This module includes the interaction between Virtual Synchronous Control (VSC) and the grid (GRID). VSC and GRID interact over a wide frequency band to generate impedance elements such as Z. 11 (jw,t) to Z nn The impedance matrix (jw,t) is ultimately represented by a Bode Diagram on the right, which shows the changes in amplitude and phase with frequency. This allows for the analysis of the characteristics of deep-sea wind power grid-connected systems through the processing of time-domain and frequency-domain information and impedance modeling.

[0101] The method provided in this application establishes a system of differential-algebraic equations containing the impedance of a large-capacity submarine cable and multiple nonlinear elements. It then utilizes the physical relationships between multiple state variables in a deep-sea wind power grid-connected control system to construct a recursive coefficient matrix. The preset global time domain is divided into multiple time domain sub-intervals. Referring to a general analytical solution expressed using power series explicit operators, the recursive coefficient matrix is ​​solved in each time domain sub-interval to generate a global time domain explicit solution corresponding to the preset global time domain. A discrete Fourier transform is performed on the global time domain explicit solution to obtain the time-frequency domain spectrum result. A time-varying impedance model is constructed based on differential algebraic equations, and the model is tuned using time-frequency domain spectral results. The tuned time-varying impedance model is then used to analyze the grid-connected control system of deep-sea wind power. By solving the full-time domain multi-segment sub-interval explicit solution and calculating the time-frequency domain spectral results, the microstep integration is significantly reduced, computational resource consumption is lowered, and computational efficiency is improved. Moreover, the explicit time-domain solution can be updated online according to the operating conditions, quickly capturing the stability changes of deep-sea grid connection, improving the characterization accuracy and adaptability, and demonstrating high applicability in the context of large-scale development of deep-sea wind power.

[0102] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application.

[0103] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenario.

[0104] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A grid-connected analysis method for time-varying impedance models of deep-sea wind power based on the time-domain residual method, characterized in that, include: A system of differential-algebraic equations incorporating the impedance of a large-capacity submarine cable and multiple nonlinear elements is established. A recursive coefficient matrix is ​​constructed using the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system. This recursive coefficient matrix describes the state change patterns of the deep-sea wind power grid-connected control system at multiple time points. Establishing the system of differential-algebraic equations includes: determining the multiple nonlinear elements, which include a PLL phase-locked loop (PLL), a droop control loop, and a filtering loop; and constructing the following PLL equations by referring to the local and global relationships of the filter in the dq coordinate system: in, It is the phase difference of the phase-locked loop. It is the local angular velocity of the inverter in the dq coordinate system. This is the global reference angular velocity of the power grid in the dq coordinate system; the following set of equations for the droop control element is constructed: in, It is active power. It is reactive power. and It is droop control gain. and It is a time constant. It is a given reference active power. Given the reference reactive power; and considering the impedance of the high-capacity submarine cable, the following set of equations for the filtering stage is constructed: in, and It is the dq component of the inverter output bus voltage. and It is the dq component of the inverter output current after passing through the filter inductor. and It is the dq component of the output grid-connected current through the grid-connected impedance. and It is the dq component of the controller output voltage. It's angular velocity. It is the filter capacitor of the filter. It is the filter inductor of the filter; the PLL phase-locked loop equation, the droop control loop equation, and the filter loop equation are combined to obtain the differential algebraic equation set; The preset global time domain is divided into multiple time domain sub-intervals, and the recursive coefficient matrix is ​​solved in each time domain sub-interval by referring to the general analytical solution represented by the power series explicit operator, thereby generating the global time domain explicit solution corresponding to the preset global time domain. The time-domain explicit solution is subjected to a discrete Fourier transform to obtain the time-frequency domain spectral results. A time-varying impedance model is constructed based on the differential algebraic equation, and the time-varying impedance model is tuned using the time-frequency domain spectral results. The tuned time-varying impedance model is then used to analyze the deep-sea wind power grid-connected control system.

2. The method according to claim 1, characterized in that, The method of constructing a recursive coefficient matrix by utilizing the physical relationships between multiple state variables in the deep-sea wind power grid-connected control system includes: Based on the relationships between the multiple state variables in the aforementioned deep-sea wind power grid-connected control system, the following full-state mapping constraint matrix is ​​derived: in, The full state quantity mapping constraint matrix is ​​used to reflect the linear and nonlinear constraint relationships between the multiple state variables. Represents the vector formed by the state variables; Referring to the full state quantity mapping constraint matrix, the following recursive coefficient matrix is ​​constructed: in, This represents the elements in the recursive coefficient matrix. Indicates a line, Indicates a column; Indicates the first Each state variable Order coefficient, The value of is a natural number; Indicates the first State variables The order coefficients are calculated using the recursive coefficient matrix and the current state variable values.

3. The method according to claim 1, characterized in that, The method refers to solving the recursive coefficient matrix in each of the time-domain sub-intervals using a general analytical solution represented by a power series explicit operator, generating a global time-domain explicit solution corresponding to the preset global time domain, including: In each of the time-domain subintervals, the general analytical solution of the time-domain subinterval is expressed using the power series explicit operator according to the following formula: in, Indicates at time State vector of the grid-connected control system for deep-sea wind power; Represents the current time domain sub-interval The initial state vector; This represents a set of explicitly expanded coefficients constructed based on the dynamic characteristics of the deep-sea wind power grid-connected control system within the current time domain sub-interval, which are recursively obtained from the dynamic equations and boundary conditions of the deep-sea wind power grid-connected control system. This indicates the current time, and the value range is within the current time domain sub-interval. Inside; Indicates the start time of the current time domain sub-interval; The Pade approximation algorithm is used to extend the convergence domain of the general analytical solution for each time domain sub-interval, resulting in the following expression. Based on this expression, the explicit time-domain power series solution for each time domain sub-interval is determined: in, Indicates at time State vector of the grid-connected control system for deep-sea wind power; It is the molecular polynomial, used to extend the approximate region of convergence; It is a denominator polynomial used to extend the approximate region of convergence; It is the numerator polynomial The coefficient; It is a denominator polynomial The coefficient; This indicates the current time, and the value range is within the current time domain sub-interval. Inside; This represents the reference time point, which serves as the benchmark point for time sampling in the Pade approximation algorithm. By defining an error loss function, the adaptive order and step size are estimated so that the explicit solution of the time-domain power series of each time-domain sub-interval remains convergent within a preset error threshold. The following formula is used to concatenate the explicit time-domain power series solutions of the multiple time-domain sub-intervals to obtain the global explicit time-domain solution: in, Represents the time in the global time domain State vector of the grid-connected control system for deep-sea wind power; This indicates a splicing operation, which splices together multiple explicit time-domain power series solutions to form the global explicit time-domain solution; Indicates the index of the time-domain sub-interval, with values ​​ranging from 1 to... , used to identify each time domain sub-interval; Represents a time-domain sub-interval Local approximate solution; The range of values ​​is within Inside; Indicates the first The start time of each time domain sub-interval Indicates the first The end time of each time domain sub-interval.

4. The method according to claim 1, characterized in that, The step of performing a discrete Fourier transform on the global time-domain explicit solution to obtain the time-frequency domain spectrum results includes: Determine a specified step size and use the specified step size as a time window; Within each time window, a discrete Fourier transform is performed on the time-domain coefficients of the global time-domain explicit solution to convert the time-domain coefficients into frequency-domain coefficients, and the time-frequency domain spectrum result is generated with reference to the frequency-domain coefficients.

5. The method according to claim 1, characterized in that, The step of tuning the time-varying impedance model using the time-frequency domain spectral results includes: A simulation environment is constructed to simulate the grid connection operation of deep-sea wind power, and characteristic frequency components in the simulation process are generated by referring to the time-frequency domain spectral results. The model parameters of the time-varying impedance model are tuned by analyzing the characteristic frequency components.

6. The method according to claim 5, characterized in that, The construction of the simulation environment, in which the grid connection operation of deep-sea wind power is simulated, and the generation of characteristic frequency components in the simulation process based on the time-frequency domain spectral results, includes: The simulation environment is constructed based on the baseline model of a single-unit infinite bus grid inverter. In the simulation environment, for the droop control inverter, a step change in the grid-connected inductance is set up, and the time-domain waveform of the transient frequency and voltage response under the simulation environment is solved by referring to the time-frequency domain waveform results. The time-domain analytical solution results are obtained, and the dynamic response data of the deep-sea wind power grid-connected control system in the simulation environment are acquired. In the hardware loop of the simulation environment, a step change in the inductive reactance of the grid-connected inductor is triggered, and the time-domain waveform of the deep-sea wind power grid-connected control system in the hardware loop is acquired in real time with reference to the time-domain analytical solution result. The dynamic response data and the time-domain waveform are subjected to spectral analysis to obtain the characteristic frequency components.

Images