New energy station grid-connected oscillation risk assessment method and device

Abstract

The invention provides a new energy station grid-connected oscillation risk assessment method and device, and relates to the field of new energy grid-connected control. The method comprises the following steps: constructing a dynamic equivalent model based on new energy station operation parameters; based on the dynamic equivalent model, obtaining impedance spectrum feature domains in different control modes; calibrating an impedance spectrum feature domain based on a semi-physical simulation platform; inputting the calibrated impedance spectrum feature domain into a grid-connected oscillation risk assessment model, and outputting oscillation damping ratio predicted values of the new energy station in different control modes based on the grid-connected oscillation risk assessment model; and constructing a self-adaptive threshold, and evaluating the grid-connected oscillation risk of the new energy station based on the magnitude relationship between the oscillation damping ratio predicted value and the self-adaptive threshold in different control modes. The method solves the problems that when different control modes are switched or external operation conditions are changed, a prediction result of a traditional method is prone to distortion, and the real oscillation risk of the new energy station is difficult to reflect accurately.

Description

Technical Field

[0001] This application relates to the field of grid connection control for new energy sources, and in particular to a method and apparatus for assessing grid connection oscillation risks of new energy power plants. Background Technology

[0002] With the large-scale integration of new energy sources into the power system, the power grid operation is showing a trend towards power electronics and low inertia, gradually weakening the traditional stability mechanisms that rely on synchronous machine inertia and damping. New energy units based on power electronic interfaces exhibit diverse control strategies and complex operating modes, leading to strong nonlinearity and time-varying electrical characteristics at the grid connection point.

[0003] In typical renewable energy power plant operation scenarios, traditional risk assessment methods based on power system equivalent impedance and empirical thresholds have significant shortcomings. Because the control parameters of renewable energy units are difficult to obtain accurately, traditional models often rely on simplified assumptions, leading to a large deviation between the established equivalent impedance and the actual operating characteristics. This is especially true when switching between different control modes or when external operating conditions change, causing the prediction results to become distorted. In other words, in this scenario, traditional methods struggle to accurately reflect the true oscillation risks of renewable energy power plants.

[0004] Therefore, there is an urgent need for a method and device for assessing the grid connection oscillation risk of new energy power plants. Summary of the Invention

[0005] This application provides a method and device for assessing the grid-connected oscillation risk of new energy power plants, which solves the problem that the prediction results of traditional methods are prone to distortion when switching between different control modes or when external operating conditions change, and thus cannot accurately reflect the real oscillation risk of new energy power plants.

[0006] The first aspect of this application provides a method for assessing the grid-connected oscillation risk of renewable energy power plants. The method includes: obtaining the operating parameters of different units in the renewable energy power plant through disturbance injection, and constructing a dynamic equivalent model based on the operating parameters; obtaining the impedance spectrum characteristic domain of the renewable energy power plant under different control modes through frequency domain analysis based on the dynamic equivalent model; constructing a hardware-in-the-loop simulation platform, and calibrating the impedance spectrum characteristic domain based on the hardware-in-the-loop simulation platform; inputting the calibrated impedance spectrum characteristic domain into the grid-connected oscillation risk assessment model, and outputting the predicted oscillation damping ratio of the renewable energy power plant under different control modes based on the grid-connected oscillation risk assessment model; constructing an adaptive threshold based on external operating characteristic parameters characterizing the power system operation mode, including grid topology parameters, power flow distribution parameters, and capacity ratio parameters; and assessing the grid-connected oscillation risk of the renewable energy power plant based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold.

[0007] Optionally, the operating parameters of different units in the new energy power station are obtained through perturbation injection, and a dynamic equivalent model is constructed based on the operating parameters. Specifically, this includes: injecting small perturbation voltage signals of preset amplitude into the ports of different units and collecting the corresponding operating parameters, including controller operating parameters, generator terminal voltage operating parameters, and generator terminal current operating parameters; constructing the frequency response characteristics of different units based on the response relationship between generator terminal voltage operating parameters and generator terminal current operating parameters; and constructing a dynamic equivalent model based on the rate response characteristics and in combination with the controller operating parameters.

[0008] Optionally, the impedance spectrum characteristic domain of the new energy power station under different control modes can be obtained through frequency domain analysis: corresponding control mode constraints are introduced into the dynamic equivalent model, including constant power control mode, voltage support control mode, and virtual synchronization control mode; under the control mode constraints, the dynamic equivalent model is subjected to small-signal linearization processing, and the linearized state equations under different control modes are obtained; based on the linearized state equations under different control modes, the impedance matrix is ​​calculated in the frequency domain, and the corresponding impedance spectrum characteristic domain is output.

[0009] Optionally, before inputting the calibrated impedance spectrum feature domain into the grid-connected oscillation risk assessment model and outputting the predicted oscillation damping ratio of new energy power plants under different control modes based on the grid-connected oscillation risk assessment model, a grid-connected oscillation risk assessment model is constructed. Specifically, this includes: extracting impedance spectrum feature domain samples based on historical operating data and historical disturbance response data; obtaining historical oscillation event label data corresponding to the impedance spectrum feature domain samples; constructing an input feature set and an output label set based on the impedance spectrum feature domain samples and historical oscillation event label data, and performing model parameter optimization training based on a training algorithm; and introducing regularization constraints and cross-validation mechanisms during the model parameter optimization training process to construct the grid-connected oscillation risk assessment model.

[0010] Optionally, the predicted oscillation damping ratio of new energy power plants under different control modes is output based on the grid-connected oscillation risk assessment model. Specifically, this includes: inputting the calibrated impedance spectrum feature domain into the grid-connected oscillation risk assessment model; performing forward inference in the grid-connected oscillation risk assessment model based on the trained input feature set and output label set; extracting the impedance spectrum feature subsets corresponding to different control modes and performing mapping operations during the forward inference process; and calculating the predicted oscillation damping ratio under each control mode based on the mapping operation results.

[0011] Optionally, an adaptive threshold is constructed based on external operating characteristic parameters characterizing the power system's operating mode. Specifically, this includes: establishing a network sensitivity matrix characterizing network connectivity and power flow transfer characteristics based on grid topology parameters; calculating the power margin index of renewable energy power plants under different operating modes based on power flow distribution parameters; quantifying the influence coefficient of renewable energy power plants on the overall grid inertia and damping levels based on capacity ratio parameters; and weighting and fusing the network sensitivity matrix, power margin index, and influence coefficients to calculate the adaptive threshold.

[0012] Optionally, the grid connection oscillation risk of new energy power plants is assessed based on the relationship between the predicted oscillation damping ratio and the adaptive threshold under different control modes. Specifically, this includes: obtaining the predicted oscillation damping ratio under different control modes and comparing it with the adaptive threshold one by one; when the predicted oscillation damping ratio is less than the adaptive threshold, it is determined that there is a grid connection oscillation risk under the corresponding control mode; when the predicted oscillation damping ratio is greater than or equal to the adaptive threshold, it is determined that there is no grid connection oscillation risk under the corresponding control mode.

[0013] A second aspect of this application provides a device for assessing the grid connection oscillation risk of a new energy power station. The device includes an acquisition module and a processing module, wherein...

[0014] The acquisition module is used to obtain the operating parameters of different units in the new energy power station through perturbation injection, and to build a dynamic equivalent model based on the operating parameters.

[0015] The processing module is used to obtain the impedance spectrum characteristic domain of new energy power plants under different control modes through frequency domain analysis based on a dynamic equivalent model; construct a hardware-in-the-loop simulation platform and calibrate the impedance spectrum characteristic domain based on the hardware-in-the-loop simulation platform; input the calibrated impedance spectrum characteristic domain into the grid-connected oscillation risk assessment model, and output the predicted oscillation damping ratio of new energy power plants under different control modes based on the grid-connected oscillation risk assessment model; construct an adaptive threshold based on external operating characteristic parameters characterizing the power system operation mode, including grid topology parameters, power flow distribution parameters, and capacity ratio parameters; and assess the grid-connected oscillation risk of new energy power plants based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold.

[0016] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described above.

[0017] A fourth aspect of this application provides a computer-readable storage medium storing a computer program, which is executed by a processor according to any of the methods described above.

[0018] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0019] 1. A dynamic equivalent model is constructed based on the operating parameters of new energy power plants. Based on the dynamic equivalent model, the impedance spectrum characteristic domain under different control modes is obtained. The impedance spectrum characteristic domain is calibrated based on a hardware-in-the-loop simulation platform. The calibrated impedance spectrum characteristic domain is input into the grid-connected oscillation risk assessment model, and the predicted oscillation damping ratio of new energy power plants under different control modes is output based on the grid-connected oscillation risk assessment model. An adaptive threshold is constructed, and the grid-connected oscillation risk of new energy power plants is assessed based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold. This allows for dynamic correction of the impedance spectrum characteristics and the generation of adaptive criteria in combination with external operating conditions under complex control modes and variable operating conditions of new energy power plants. This ensures that the risk assessment results are consistent with the actual operating state, solving the technical problems of large modeling deviations due to the unavailability of unit parameters and insufficient assessment accuracy due to rigid fixed thresholds in traditional methods. This improves the accuracy and adaptability of grid-connected oscillation risk assessment for new energy power plants.

[0020] 2. By injecting small disturbance voltage signals of preset amplitude into the ports of different generating units and collecting corresponding operating parameters, including controller operating parameters, generator terminal voltage operating parameters, and generator terminal current operating parameters, frequency response characteristics of different generating units are constructed based on the response relationship between generator terminal voltage operating parameters and generator terminal current operating parameters. Based on the rate response characteristics and combined with controller operating parameters, a dynamic equivalent model is constructed, which can accurately characterize the dynamic characteristics of various types of generating units in new energy power plants under different operating conditions. This solves the technical problem of large deviations in equivalent models caused by the unavailability of internal control parameters of the generating units or the failure to dynamically identify generator terminal characteristics in traditional methods, and provides a reliable modeling foundation for subsequent impedance spectrum characteristic domain acquisition and risk assessment.

[0021] 3. Corresponding control mode constraints are introduced into the dynamic equivalent model. These control modes include constant power control, voltage support control, and virtual synchronization control. Under these constraints, the dynamic equivalent model undergoes small-signal linearization to obtain linearized state equations for different control modes. Based on these linearized state equations, the impedance matrix is ​​calculated in the frequency domain, and the corresponding impedance spectrum feature domain is output. This comprehensively reflects the dynamic characteristics of the port impedance of new energy power stations under different control strategies, solving the technical problem that traditional methods can only model a single control mode and cannot cover multi-mode switching scenarios. This provides reliable feature inputs for subsequent parameter identification and calibration, and grid connection oscillation risk prediction. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating a method for assessing grid-connected oscillation risk at a new energy power station, as provided in an embodiment of this application.

[0023] Figure 2 This is a schematic diagram of a module of a new energy power station grid connection oscillation risk assessment device provided in an embodiment of this application;

[0024] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0025] Explanation of reference numerals in the attached figures: 21. Acquisition module; 22. Processing module; 301. Processor; 302. Communication bus; 303. User interface; 304. Network interface; 305. Memory. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0027] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0028] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0029] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0030] Please refer to Figure 1 The diagram shows a flowchart of a method for assessing grid-connected oscillation risk of a new energy power station provided in an embodiment of this application. The flowchart mainly includes the following steps: S101 to S106.

[0031] Step S101: Obtain the operating parameters of different units in the new energy power station through perturbation injection, and construct a dynamic equivalent model based on the operating parameters.

[0032] Specifically, new energy power plants typically include multiple types of grid-connected units, such as wind turbine generators, photovoltaic inverter units, and energy storage converter units. These different types of units differ significantly in their topology, grid connection point electrical characteristics, and internal control strategies. For example, wind turbine generators usually rely on doubly-fed or full-power converter control, photovoltaic inverter units are based on DC-AC conversion, and energy storage converter units need to consider both charging and discharging bidirectional power regulation. Because these units are affected by external grid disturbances and internal controller settings during grid-connected operation, their electrical and control behaviors are time-varying, making it difficult to accurately characterize their dynamic characteristics with a single static parameter.

[0033] Therefore, in the operation scenario of new energy power plants, it is necessary to apply a small disturbance voltage signal of preset amplitude at the unit port to stimulate the voltage-current coupling characteristics of the unit under different operating conditions and collect the resulting operating parameters. These operating parameters cover the operating parameters at the unit controller level, as well as the response data of the terminal voltage and current, thus reflecting the dynamic characteristics of the unit under disturbance. By processing the operating parameters, a set of dynamic characteristics for various units can be obtained, and a unified dynamic equivalent model can be further constructed to map the overall multi-unit characteristics of the new energy power plant into an analyzable unified mathematical model, laying the foundation for subsequent frequency domain analysis, parameter identification, and oscillation risk assessment.

[0034] In one possible implementation, step S101 further includes: obtaining the operating parameters of different units in the new energy power station through a disturbance injection method, and constructing a dynamic equivalent model based on the operating parameters. Specifically, this includes: injecting small disturbance voltage signals of preset amplitude into the ports of different units and collecting the corresponding operating parameters, including controller operating parameters, generator terminal voltage operating parameters, and generator terminal current operating parameters; constructing the frequency response characteristics of different units based on the response relationship between the generator terminal voltage operating parameters and the generator terminal current operating parameters; and constructing a dynamic equivalent model based on the rate response characteristics and in combination with the controller operating parameters.

[0035] Specifically, a constraint description is provided for typical grid-connected scenarios of new energy power plants, and a disturbance injection and parameter acquisition process is organized accordingly. New energy power plants include various types of units such as wind turbine generators, photovoltaic inverter units, and energy storage converter units. Different units differ in control strategies, grid-connected electrical interfaces, and dynamic response speeds. Therefore, without altering the grid-connected safety boundary, small-signal disturbances are applied to the ports of each unit to induce dynamic coupling between port voltage and port current. Disturbance injection employs amplitude-limited swept-frequency signals or pseudo-random binary sequences, constraining the maximum amplitude relative to the rated phasor to not exceed a preset percentage to avoid triggering protection actions. Disturbance metadata is recorded, including disturbance amplitude, swept-frequency path, start and end frequencies, step resolution, and sampling frequency. Operating parameters are synchronously acquired after the unit's local clock is aligned with the power plant's time reference. In this implementation, the operating parameters are strictly categorized into two main types: controller operating parameters and terminal electrical operating parameters. Controller operating parameters characterize the real-time setting and limiting states of the inner and outer loops, including current loop proportional and integral coefficients, voltage loop proportional and integral coefficients, reactive power or voltage targets, virtual inertia coefficients, virtual damping coefficients, current and voltage limiting thresholds, sampling period, and modulation strategy identifiers. Terminal electrical operating parameters characterize the observable external characteristics of the grid-connected port, including three-phase voltage phasors and three-phase current phasors, symmetrical component decomposition results, port phase angle difference, harmonic spectrum amplitude and phase, short-time voltage change rate, and DC component indices. Simultaneously, the grid-connected point frequency and short-circuit ratio estimates are recorded for consistency verification. To ensure data quality, the synchronized operating parameters undergo denoising, drift reduction, and window function shaping. Phasor reconstruction and synchronization averaging are used to suppress background disturbances, and outlier detection is used to remove data segments during saturation and unlocking periods, thus outputting a dataset suitable for frequency domain identification.

[0036] Based on the dataset, frequency response characteristics of different units are constructed. Specifically, the equivalent impedance of the port is estimated at each disturbance frequency point, and a frequency response curve is generated. The formula for defining the equivalent impedance of the port is: Where Z(ω) is the frequency domain function of the port equivalent impedance, used to characterize the voltage-current relationship at the port at angular frequency ω; V(ω) is the complex spectral phasor of the terminal voltage at angular frequency ω, used to represent the voltage amplitude and phase at that frequency; I(ω) is the complex spectral phasor of the terminal current at angular frequency ω, used to represent the current amplitude and phase at that frequency; ω is the angular frequency parameter, the value of which is given by the perturbation sweep path. To adapt to the coupling characteristics of the power electronic unit in the synchronous rotating coordinate system, the three-phase quantities are mapped to dq coordinates and the matrix frequency response characteristics of the port admittance or impedance are extracted. Through vector fitting and rational function modeling, a stable and causal transfer function expression within the target frequency band is obtained, and the response samples are archived with control mode labels. The control modes include at least constant power control mode, voltage support control mode, and virtual synchronous control mode. The frequency response characteristics are accepted as the nominal external characteristics of the unit after amplitude-phase consistency test and residual threshold test, and used for the next step of model construction.

[0037] In terms of model construction, this implementation uses controller operating parameters as prior constraints and verified frequency response characteristics as fitting targets. A dynamic equivalent model is formed by combining a gain-scheduled port equivalent model with a state-space model. Specifically, a segmented model library driven by multi-condition labels and cross-segment continuity constraints jointly ensure availability under control mode switching and setpoint drift conditions. Regularized least-squares or Bayesian update strategies are used to synchronously estimate control-related effective parameters, equivalent RLC, and time delay elements to match observed amplitude and phase characteristics. Simultaneously, cross-validation is performed by independently reserving disturbance segments to control overfitting risk. If the validation residual exceeds a threshold, re-injection and re-estimation are triggered to achieve closed-loop convergence. For multiple units of the same type, parameter sharing and individual fine-tuning are first performed within the homogeneous subgroup to reduce identification variance. Then, at the grid connection point, a site-level aggregated dynamic equivalent model is obtained based on the multi-port network equivalence principle.

[0038] Step S102: Based on the dynamic equivalent model, the impedance spectrum characteristic domain of the new energy power station under different control modes is obtained through frequency domain analysis.

[0039] Specifically, the frequency domain analysis uses the dynamic equivalent model output in step S101 as input. For the operating conditions of the renewable energy power plant under different control modes such as constant power control, voltage support control, and virtual synchronization control, small-signal disturbance analysis and linearization processing are performed on the model. The amplitude-phase relationship between port voltage and port current at different frequencies is extracted, and the distribution characteristics of the port equivalent impedance in the frequency domain are calculated accordingly. Through point-by-point analysis and fitting within a preset frequency range, an impedance spectrum characteristic domain that can characterize the overall dynamic characteristics of the renewable energy power plant is formed. This domain is used to describe the stability boundary of the interaction between the renewable energy power plant and the power system under different control modes, providing basic data for subsequent parameter identification and oscillation risk assessment.

[0040] In one possible implementation, step S102 further includes: introducing corresponding control mode constraints into the dynamic equivalent model, the control modes including constant power control mode, voltage support control mode and virtual synchronization control mode; under the control mode constraints, performing small-signal linearization processing on the dynamic equivalent model and obtaining linearized state equations under different control modes; based on the linearized state equations under different control modes, calculating the impedance matrix in the frequency domain and outputting the corresponding impedance spectrum feature domain.

[0041] Specifically, after introducing constraints for constant power control mode, voltage support control mode, and virtual synchronous control mode into the dynamic equivalent model, small-signal linearization is performed at the operating points of each control mode to obtain the port small-signal model. The three-phase time-domain port quantities are mapped to a synchronous rotating coordinate system to decouple the positive-sequence components and cross-coupling terms, and the frequency response of the port impedance matrix is ​​solved point-by-point within a given frequency set. The port impedance matrices obtained under each control mode form a continuous trajectory on the frequency axis, and the set of trajectories of the impedance matrix eigenvalues ​​in the complex plane is used as the definition of the impedance spectrum characteristic domain. This impedance spectrum characteristic domain is used to characterize the amplitude and phase characteristics and coupling strength of the new energy power station ports in the frequency domain under different control modes, serving as the processing input for subsequent parameter identification and calibration and grid connection oscillation risk assessment.

[0042] In the mapping process from three-phase quantities to the synchronous rotating coordinate system, a synchronous transformation matrix is ​​used to obtain the port quantities of the synchronous rotating coordinate system. The specific expression is as follows:

[0043] x dq =T(θ)x abc

[0044]

[0045] Where, x abc Let x be the generalized variable vector for the three-phase ports. dq Let T(θ) be the generalized variable vector of the port of the synchronous rotating coordinate system. Let T(θ) be the synchronous transformation matrix. Let θ be the electrical angle of the synchronous rotating coordinate system. The small-signal linearized expression of the dynamic equivalent model output from step S101 under control mode constraints is also given. for:

[0046]

[0047] Where Δx is the increment of the small-signal state vector. Δv dq Δi represents the small-signal vector increment of the port voltage in the synchronously rotating coordinate system. dq Let A be the small-signal vector increment of the port current in the synchronously rotating coordinate system. Let B be the small-signal state matrix. v C is the voltage-to-state input matrix. D is the state-to-current output matrix.v This is the direct voltage-to-current matrix. From this, the frequency response Y of the port admittance matrix can be obtained. dq (jω) and the frequency response Z of the port impedance matrix dq (jω) are respectively:

[0048] Y dq (jω)=C(jωI-A) -1 B v +D v

[0049] Z dq (jω)=(Y dq (jω)) -1

[0050] Where I is the identity matrix of the same order as A. j is the imaginary unit. ω is the angular frequency. Y dq (jω) is the frequency response of the port admittance matrix of the synchronous rotating coordinate system. dq (jω) represents the frequency response of the port impedance matrix in the synchronous rotating coordinate system. To explicitly characterize the influence of the control mode on the system matrix, the system matrix is ​​written in affine decomposition form based on the mode parameters:

[0051]

[0052] And for each set of mode parameters, the set of eigenvalues ​​of the impedance matrix is ​​obtained over the frequency set:

[0053]

[0054] Simultaneously define the amplitude and phase trajectories as follows:

[0055]

[0056] Will The continuous trajectory on the ω-axis serves as an equivalent representation of the impedance spectrum characteristic domain in polar coordinates. To improve numerical stability and physical consistency, at each frequency point (jωI-A)... -1 By fitting rational functions and applying non-positive constraints on the real parts of the poles, the obtained Y is guaranteed. dq Strict causality and limited passive approximation error. Here, A0 is the reference state matrix without control mode adjustable factors. B0 is the reference input matrix without control mode adjustable factors. D0 is the reference pass-through matrix without control mode adjustable factors. M k N is the mode sensitivity basis matrix of the state matrix. k P is the mode sensitivity basis matrix of the input matrix. k α is the mode sensitivity basis matrix of the pass-through matrix. p is the number of sensitivity basis matrices used for expansion. k To act on Mk The mode weighting coefficient. β k For action on N k The mode weight coefficient. γ k To act on P k The mode weight coefficient. The superscript c is the control mode index. This represents the port impedance matrix corresponding to the control mode. Ω is the discrete or continuous set for frequency analysis. λ i (·) represents the matrix eigenvalue operator. i is the eigenvalue index. This represents the amplitude of the corresponding eigenvalue at angular frequency ω. This represents the phase of the corresponding eigenvalue at angular frequency ω. The set of continuous trajectories of the above frequency domain results constitutes the impedance spectrum characteristic domain under this control mode.

[0057] Step S103: Construct a hardware-in-the-loop simulation platform and calibrate the impedance spectrum characteristic domain based on the hardware-in-the-loop simulation platform.

[0058] Specifically, by introducing a dynamic equivalent model of a new energy power plant into a hardware-in-the-loop simulation platform and combining it with actually collected operating parameters, key controller parameters, electrical interface parameters, and equivalent impedance parameters in the model are identified. The hardware-in-the-loop simulation platform can establish interaction between the real controller hardware and the simulated power grid environment, enabling the online generation of port response data that is highly consistent with actual operation under different operating conditions and control modes. The specific process is as follows:

[0059] First, under the operating condition with control mode index c, taking the port small-signal model of the synchronous rotating coordinate system as the baseline, the linearized model is denoted as:

[0060]

[0061] And calculate the model admittance in the frequency domain and model impedance Where Δx is the increment of the small-signal state vector. Its time derivative increment, Δv dq The small-signal vector increment of the port voltage, Δi dq For the small-signal vector increment of the port current, A c The state matrix, Given the input matrix, C c For the output matrix, Let I be a through matrix, and A be a matrix with the same name as A. c An identity matrix of the same order, where j is the imaginary unit, ω is the angular frequency, and c is the control mode index. Simultaneously, a small, amplitude-limited perturbation is injected into the semi-physical platform with the same sweep frequency set Ω, and the measurement admittance is calculated:

[0062]

[0063] And based on this, the measured impedance is:

[0064]

[0065] in, Let be the complex phasor of the port voltage at angular frequency ω. Let Ω be the complex phasor of the port current at angular frequency ω, and Ω be the set of sweep frequencies. Next, a parameter identification vector is constructed, and a weighted residual minimization problem is established. Let the parameter vector to be identified be...

[0066]

[0067] Using the Frobenius norm to measure the frequency domain residuals, the objective function is defined. for:

[0068]

[0069]

[0070] Where w(ω) is the frequency weight, ||·|| F Let λ be the Frobenius norm, and λ be the regularization coefficient. Let J be the prior parameter, and J be the pair of parameters. Regarding θ c Jacobi, J * For the conjugate transpose, W = diag({w(ω)I4} ω∈Ω ) is a weighted matrix, I θ For θ c The identity matrix is ​​of the same order, and r is the vectorized result of the complex matrix residuals stacked in Ω. Next, physical feasibility and passive projection are introduced to constrain the calibration results. A numerical approximation of the necessary and sufficient condition for port passivity is applied to each ω, making the real admittance positive semi-definite:

[0071]

[0072] Where Sym(·) is the Hermitian symmetry operator, (·) * For the conjugate transpose, ≥0 indicates a positive semidefinite constraint. To suppress high-frequency noise and ensure causality, for (jωI-A) c ) -1 Perform rational function fitting and add constraints on the non-positive real parts of the poles as follows: Where σ(·) is the spectral set operator, For real part operators.

[0073] Subsequently, after convergence was completed in control mode c, the calibration parameters were used. The admittance and impedance are reconstructed, and their continuous trajectory over Ω is used as the characteristic domain of the calibration impedance spectrum under this control mode. The above identification and projection are repeated for the entire set of control modes, ultimately yielding multi-mode calibration results covering "constant power control mode, voltage-supported control mode, and virtual synchronization control mode". The eigenvalues ​​of the calibration impedance matrix are extracted and the amplitude-phase trajectory is generated.

[0074]

[0075] Where, λ i (·) represents the eigenvalue operator. The amplitude trajectory of the i-th eigenvalue. For phase trajectory. The set of trajectories is used as the polar coordinate representation of the calibrated impedance spectrum characteristic domain, and the... This serves as the input for the subsequent "grid-connected oscillation risk assessment model".

[0076] Step S104: Input the calibrated impedance spectrum characteristic domain into the grid-connected oscillation risk assessment model, and output the predicted value of the oscillation damping ratio of the new energy power station under different control modes based on the grid-connected oscillation risk assessment model.

[0077] Specifically, the calibrated impedance spectrum feature domain is fed as a feature input into the grid-connected oscillation risk assessment model. This model is a nonlinear prediction model trained based on historical operating data and historical oscillation event data, capable of pattern recognition and mapping of impedance features under different control modes. Internally, the model performs dimensionality reduction and principal feature decomposition on the impedance spectrum feature domain through a feature extraction layer, and then combines external operating feature parameters to weight and correct the input features, establishing a mapping relationship between the input features and the oscillation mode damping ratio. After forward inference of the training parameters, the model can output corresponding predicted oscillation damping ratio values ​​under constant power control mode, voltage support control mode, and virtual synchronization control mode, respectively. The predicted oscillation damping ratio is used to characterize the potential oscillation attenuation capability or divergence trend of new energy power plants under different control modes, and serves as a basic indicator for subsequent risk assessment and threshold comparison.

[0078] In one possible implementation, step S104 further includes: extracting impedance spectrum feature domain samples based on historical operating data and historical disturbance response data; obtaining historical oscillation event label data corresponding to the impedance spectrum feature domain samples; constructing an input feature set and an output label set based on the impedance spectrum feature domain samples and historical oscillation event label data, and performing model parameter optimization training based on a training algorithm; introducing regularization constraints and cross-validation mechanisms during the model parameter optimization training process to construct a grid-connected oscillation risk assessment model; inputting the calibrated impedance spectrum feature domain into the grid-connected oscillation risk assessment model; performing forward inference in the grid-connected oscillation risk assessment model based on the trained input feature set and output label set; extracting impedance spectrum feature subsets corresponding to different control modes and performing mapping operations during the forward inference process; and calculating the predicted oscillation damping ratio under each control mode based on the mapping operation results.

[0079] Specifically, when extracting impedance spectrum feature domain samples based on historical operating data and historical disturbance response data, the control mode index of each historical sample is used as a constraint. The port impedance matrix of each historical sample is represented as follows: or its calibration form Extract the matrix eigenvalue sequence from the frequency set Ω. To form a learnable set of input features, the amplitude and phase are stacked at discrete frequency points to form a feature vector, specifically defined as:

[0080]

[0081] in, Let be the amplitude of the i-th eigenvalue at angular frequency ω. For the corresponding phase, i is the eigenvalue index, and c is the control mode index. This is a set of frequencies. To eliminate dimensional differences and magnitude bias, the features are standardized:

[0082]

[0083] in, The standardized feature vector is given by μ, which is the mean vector obtained statistically from the training set, and σ, which is the standard deviation vector obtained statistically from the training set. When obtaining the historical oscillation event label data corresponding to the impedance spectrum feature domain samples, the damping ratio is calculated using the measured oscillation waveforms of the historical events and used as the output label set. Labels are constructed at discrete peak points using the logarithmic decay method, defining the logarithmic decay δ when adjacent peak amplitudes are A1 and A2:

[0084]

[0085] Therefore, the damping ratio ζ can be calculated:

[0086]

[0087] Where δ is the logarithmic decay parameter, used to characterize the decay intensity of adjacent peak amplitudes; A1 is the first peak amplitude parameter; A2 is the second peak amplitude parameter; ζ is the damping ratio label parameter, the smaller the value, the weaker the oscillation decay capability. ζ is calculated for each historical sample under different control modes. c And form an output label set. After constructing the input feature set and output label set based on impedance spectrum feature domain samples and historical oscillation event label data, a supervised learning paradigm is used to train the grid-connected oscillation risk assessment model. Let the model mapping be f. θ (·), whose parameter vector is θ, and the training objective uses weighted Huber loss with an l2 regularization term:

[0088]

[0089] in, α is the overall loss function parameter; N is the number of training samples; n,c The sample weight parameters for the nth sample under control mode c; Here are the parameters for the Huber loss function, δ h For Huber threshold parameters; The model's predicted output parameters for the control mode c of the nth sample; Let be the corresponding damping ratio label parameter; λ be the regularization coefficient parameter; and ||θ||2 be the L2 norm parameter of the parameter vector. To ensure generalization ability and robustness, K-fold cross-validation is used, with validation set loss. Optimal is used as the early stopping criterion, and K is the fold parameter. To validate the loss parameters.

[0090] When inputting the calibrated impedance spectrum feature domain into the grid-connected oscillation risk assessment model, a selection matrix is ​​first constructed according to the control mode index to extract a subset of the input features, defined as follows:

[0091]

[0092] in, S is the feature subset parameter input to the model. c The parameters of the binary selection matrix corresponding to control mode c are: the number of rows equal to the required feature dimension, and the number of columns equal to... The dimension. The model performs forward inference to obtain the predicted oscillation damping ratio for each control mode:

[0093]

[0094] in, Let θ be the predicted oscillation damping ratio parameter for control mode c.＊ These are the optimal parameter vectors determined through training and cross-validation. To enhance physical consistency, monotonicity and range constraints are added to the output layer, mapping the predictions to a limited interval.

[0095]

[0096] Wherein, σ(·) is a Sigmoid-type compression mapping parameter, used to restrict the predicted value to a closed interval between zero and one. Thus, the predicted oscillation damping ratios for constant power control mode, voltage support control mode, and virtual synchronization control mode are obtained respectively, and used as inputs for subsequent processing steps.

[0097] Step S105: Construct an adaptive threshold based on external operating characteristic parameters that characterize the power system operation mode. The external operating characteristic parameters include grid topology parameters, power flow distribution parameters, and capacity ratio parameters.

[0098] Specifically, under different operating scenarios, the network structure, power flow distribution, and the output ratio of renewable energy plants in a power system significantly affect the overall damping level and stability margin of the system. By acquiring grid topology parameters characterizing the connectivity of grid nodes, combining them with power flow distribution parameters reflecting power transfer and distribution, and capacity ratio parameters quantifying the changes in the proportion of renewable energy installed capacity and inertia level, an external feature set capable of dynamically characterizing the power system's operating mode is established. Based on this external feature set, a comprehensive stability index is generated using pre-defined weighted fusion rules or empirical functions, and this comprehensive stability index is mapped to a corresponding adaptive threshold. This adaptive threshold can be dynamically adjusted according to changes in the power system's operating mode and is used to compare with the predicted oscillation damping ratio, thereby achieving adaptive assessment of grid-connected oscillation risk under different operating conditions.

[0099] In one possible implementation, step S105 further includes: establishing a network sensitivity matrix characterizing network connectivity and power flow transfer characteristics based on power grid topology parameters; calculating the power margin index of new energy power plants under different operating modes based on power flow distribution parameters; quantifying the influence coefficient of new energy power plants on the overall power grid inertia level and damping level based on capacity ratio parameters; and weighting and fusing the network sensitivity matrix, power margin index, and influence coefficient to calculate an adaptive threshold.

[0100] Specifically, a network sensitivity matrix is ​​established to characterize the grid topology and power flow transfer characteristics, and a frequency-domain independent sensitivity profile is constructed using the power transfer distribution factor under DC power flow.

[0101]

[0102] Where A is the line-to-node correlation matrix, used to describe the connectivity of the power grid topology; This is the diagonal matrix of line susceptance. B is the scalar capacitance of each line; bus T is the nodal susceptance matrix; PTDF This is the power transfer distribution factor matrix, used to measure the impact of unit injection on the power flow of each line; The weight matrix is ​​normalized according to the thermal stability limit of the line. S is the rated allowable power flow amplitude of the line; net This is the network sensitivity matrix, used to simultaneously reflect the topological connectivity and the power flow sensitivity of restricted lines. To obtain a single scalar index characterizing network sensitivity, a network sensitivity normalization index is defined:

[0103]

[0104] Where ||·||2 is the spectral norm, and к net The network sensitivity normalization index ranges from zero to one; a larger value indicates stronger connectivity and greater sensitivity of restricted lines to injection. Secondly, the power margin index of relevant paths to renewable energy power plants is calculated to quantify the intensity of power flow distribution.

[0105]

[0106] in, For the line The actual tidal current amplitude; For the line The single-line power margin; the larger the value, the more sufficient the margin. A collection of connecting lines electrically connected to the new energy power station; μ P This represents the minimum power margin from the perspective of connectivity constraints, used to characterize the weakest link in the power transmission corridor from the station. To ensure consistency with subsequent weighting, its stress normalization is defined as follows:

[0107] χ P =1-μ P

[0108] Where, χ P This is a power margin tension index, ranging from zero to one; a larger value indicates that the channel is closer to saturation. Next, the influence coefficient of the quantified capacity ratio on the system's equivalent inertia and equivalent damping levels is measured using a combination of capacity ratio and primary frequency regulation capability.

[0109]

[0110] Where ρ represents the proportion of new energy capacity; P RE The active power rating of new energy installed capacity; PSG The active power rating of the synchronous generator's installed capacity; For a set of synchronizers, A collection of new energy generating units; H g Let g be the unit inertia constant of the synchronous machine; The equivalent value of the virtual inertia constant provided by the new energy unit r under grid-connected control; S g With S r These are the rated capacities of the corresponding units; R g Let g be the droop coefficient of the primary frequency modulation of the synchronous machine. The equivalent value of the virtual droop coefficient for new energy generating units; H eff The capacity-weighted average equivalent inertia; D eff The capacity-weighted average equivalent damping or primary frequency modulation response strength; H ref With D ref The equivalent inertia and equivalent damping are calibrated under reference operating conditions; η is the influence coefficient of the capacity ratio on the combined effect of inertia and damping, ranging from zero to one. A larger value indicates that the overall inertia and damping are more conducive to stability. To unify the direction, the unfavorable degree χ is defined. η for:

[0111] χ η =1-η,

[0112] Where, χ η The inertia-damping disadvantage index ranges from zero to one, with larger values ​​indicating insufficient inertia and damping. Finally, the three types of quantification results are weighted and fused to calculate an adaptive threshold, first performing scale alignment and then mapping it to a damping ratio threshold.

[0113] Γ=w1·к net +w2·χ P +w3·χ η

[0114] ζ thr =ζ0·(1+α·Γ)

[0115] Where Γ is the comprehensive risk score of external operation mode, with a value range of zero to one; w1, w2, w3 are non-negative weighting coefficients, satisfying w1+w2+w3=1, used to allocate weights among network sensitivity, power margin tension, and inertia damping adverseness; ζ thr Γ is the adaptive threshold, used as the criterion for the oscillation damping ratio; ζ0 is the baseline threshold, used to calibrate the empirical lower limit; α is the threshold amplification factor, used to adjust the sensitivity of the threshold to external features. The above two equations ensure that when the network is more sensitive, the channels are more strained, and the inertia damping is less sufficient, Γ increases, thus ζ... thr Shifting upwards makes the criteria more stringent; when the operating environment is more lenient, Γ decreases, and ζ... thr It fell back to near ζ0.

[0116] The adaptive threshold ζ is obtained through the above process. thr It relies solely on external operating characteristic parameters and does not form a cyclic dependency with the predicted value of the oscillation damping ratio in step S104. It can be updated synchronously with changes in grid topology, power flow shifts, and dynamic changes in capacity ratio. It is then used as the processing input for risk assessment in step S106 and compared with the predicted value of the oscillation damping ratio, thereby achieving adaptive grid oscillation risk determination.

[0117] Step S106: Based on the relationship between the predicted value of the oscillation damping ratio and the adaptive threshold under different control modes, assess the grid connection oscillation risk of the new energy power station.

[0118] Specifically, based on the adaptive threshold ζ thr When the predicted oscillation damping ratio under a certain control mode is less than the adaptive threshold, it is determined that there is a risk of grid-connected oscillation under that control mode; when the predicted oscillation damping ratio is greater than or equal to the adaptive threshold, it is determined that the grid-connected operation under that control mode is stable. Furthermore, the judgment results from all control modes are merged to generate an overall grid-connected oscillation risk assessment conclusion for the renewable energy power plants, and corresponding levels can be output according to the severity of the risk, providing a reference for grid dispatching and renewable energy power plant control strategies.

[0119] In one possible implementation, step S106 further includes: obtaining the predicted value of the oscillation damping ratio under different control modes and comparing it with the adaptive threshold one by one; when the predicted value of the oscillation damping ratio is less than the adaptive threshold, it is determined that there is a risk of grid oscillation under the corresponding control mode; when the predicted value of the oscillation damping ratio is greater than or equal to the adaptive threshold, it is determined that there is no risk of grid oscillation under the corresponding control mode.

[0120] Specifically, the grid-connected oscillation risk assessment model, constructed based on historical operating data and historical disturbance response data, has output predicted values ​​of the oscillation damping ratio under constant power control mode, voltage support control mode, and virtual synchronous control mode, denoted as follows: and In step S105, the external operation mode index, which is formed by combining power grid topology parameters, power flow distribution parameters, and capacity ratio parameters, is weighted and fused to obtain the adaptive threshold ζ. thr This threshold is dynamically adjusted according to changes in the power grid's operating mode.

[0121] In this embodiment, the predicted oscillation damping ratio for each control mode is compared with the adaptive threshold one by one. For example, in constant power control mode, if This indicates that under this mode, the oscillation damping capability of new energy power stations is insufficient when subjected to disturbances, posing a risk of grid connection oscillation; if ζthr This indicates that the operation in this mode is stable. Performing the same comparison in the voltage support control mode, if... If a risk exists in this mode, the output will be given; otherwise, it will be considered stable. In virtual synchronization control mode, if... This indicates that the virtual inertia and damping of the new energy power station under this mode are insufficient to guarantee system stability, and should also be judged as having a risk; if This mode will then operate stably.

[0122] Please refer to Figure 2 This document illustrates a schematic diagram of a grid-connected oscillation risk assessment device for a new energy power station, provided in an embodiment of this application. The device includes an acquisition module 21 and a processing module 22.

[0123] The acquisition module 21 is used to acquire the operating parameters of different units in the new energy power station through perturbation injection, and to construct a dynamic equivalent model based on the operating parameters.

[0124] Processing module 22 is used to obtain the impedance spectrum characteristic domain of new energy power plants under different control modes through frequency domain analysis based on a dynamic equivalent model; construct a hardware-in-the-loop simulation platform and calibrate the impedance spectrum characteristic domain based on the hardware-in-the-loop simulation platform; input the calibrated impedance spectrum characteristic domain into the grid-connected oscillation risk assessment model, and output the predicted oscillation damping ratio of new energy power plants under different control modes based on the grid-connected oscillation risk assessment model; construct an adaptive threshold based on external operating characteristic parameters characterizing the power system operation mode, including grid topology parameters, power flow distribution parameters, and capacity ratio parameters; and assess the grid-connected oscillation risk of new energy power plants based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold.

[0125] In one possible implementation, the acquisition module 21 is used to acquire the operating parameters of different units in the new energy power station through disturbance injection, and to construct a dynamic equivalent model based on the operating parameters. Specifically, this includes: injecting small disturbance voltage signals of preset amplitude into the ports of different units and collecting the corresponding operating parameters, including controller operating parameters, generator terminal voltage operating parameters, and generator terminal current operating parameters; constructing the frequency response characteristics of different units based on the response relationship between the generator terminal voltage operating parameters and the generator terminal current operating parameters; and constructing a dynamic equivalent model based on the rate response characteristics and in combination with the controller operating parameters.

[0126] In one possible implementation, the processing module 22 is used to obtain the impedance spectrum characteristic domain of the new energy power station under different control modes through frequency domain analysis: introducing corresponding control mode constraints into the dynamic equivalent model, the control modes including constant power control mode, voltage support control mode and virtual synchronization control mode; under the control mode constraints, performing small-signal linearization processing on the dynamic equivalent model and obtaining the linearized state equations under different control modes; based on the linearized state equations under different control modes, calculating the impedance matrix in the frequency domain and outputting the corresponding impedance spectrum characteristic domain.

[0127] In one possible implementation, the processing module 22 is used to construct a grid-connected oscillation risk assessment model before inputting the calibrated impedance spectrum feature domain into the grid-connected oscillation risk assessment model and outputting the predicted oscillation damping ratio of the new energy power station under different control modes based on the grid-connected oscillation risk assessment model. Specifically, this includes: extracting impedance spectrum feature domain samples based on historical operating data and historical disturbance response data; obtaining historical oscillation event label data corresponding to the impedance spectrum feature domain samples; constructing an input feature set and an output label set based on the impedance spectrum feature domain samples and historical oscillation event label data, and performing model parameter optimization training based on a training algorithm; and introducing regularization constraints and cross-validation mechanisms during the model parameter optimization training process to construct the grid-connected oscillation risk assessment model.

[0128] In one possible implementation, the processing module 22 is used to output the predicted oscillation damping ratio of new energy power plants under different control modes based on the grid-connected oscillation risk assessment model. Specifically, it includes: inputting the calibrated impedance spectrum feature domain into the grid-connected oscillation risk assessment model; performing forward inference in the grid-connected oscillation risk assessment model based on the trained input feature set and output label set; extracting the impedance spectrum feature subsets corresponding to different control modes and performing mapping operations during the forward inference process; and calculating the predicted oscillation damping ratio under each control mode based on the mapping operation results.

[0129] In one possible implementation, the processing module 22 is used to construct an adaptive threshold based on external operating characteristic parameters characterizing the power system's operating mode. Specifically, this includes: establishing a network sensitivity matrix characterizing network connectivity and power flow transfer characteristics based on grid topology parameters; calculating the power margin index of renewable energy power plants under different operating modes based on power flow distribution parameters; quantifying the influence coefficient of renewable energy power plants on the overall grid inertia and damping levels based on capacity ratio parameters; and weightedly fusing the network sensitivity matrix, power margin index, and influence coefficients to calculate the adaptive threshold.

[0130] In one possible implementation, the processing module 22 is used to assess the grid connection oscillation risk of new energy power plants based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold. Specifically, this includes: obtaining the predicted oscillation damping ratio under different control modes and comparing it with the adaptive threshold one by one; when the predicted oscillation damping ratio is less than the adaptive threshold, determining that there is a grid connection oscillation risk under the corresponding control mode; when the predicted oscillation damping ratio is greater than or equal to the adaptive threshold, determining that there is no grid connection oscillation risk under the corresponding control mode.

[0131] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0132] This application also provides an electronic device. (See reference...) Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: at least one processor 301, at least one communication bus 302, a user interface 303, at least one network interface 304, and a memory 305.

[0133] The communication bus 302 is used to enable communication between these components.

[0134] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0135] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0136] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling data stored in the memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0137] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. (Refer to...) Figure 3 The memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application for assessing the risk of grid connection oscillations in new energy power plants.

[0138] exist Figure 3In the illustrated electronic device, the user interface 303 is primarily used to provide an input interface for the user and acquire user input data; while the processor 301 can be used to call the application program for assessing grid-connected oscillation risks of new energy power plants stored in the memory 305. When executed by one or more processors 301, the electronic device performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0139] This application also provides a computer-readable storage medium storing instructions. When executed by one or more processors, these instructions cause an electronic device to perform one or more of the methods described in the above embodiments.

[0140] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0141] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0142] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0143] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0144] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0145] The above description is merely an exemplary embodiment disclosed in this application and should not be construed as limiting the scope of this application. Any equivalent changes and modifications made in accordance with the teachings of this application shall still fall within the scope of this application.

[0146] This application is intended to cover any variations, uses, or adaptations disclosed herein that follow the general principles disclosed herein and include common knowledge or customary technical means in the art that are not described in this application.

Claims

1. A method for assessing the risk of grid-connected oscillations at renewable energy power plants, characterized in that, The method includes: The operating parameters of different units in the new energy power station are obtained by perturbation injection, and a dynamic equivalent model is constructed based on the operating parameters; Based on the dynamic equivalent model, the impedance spectrum characteristic domain of the new energy power station under different control modes is obtained through frequency domain analysis. A hardware-in-the-loop simulation platform is constructed, and the impedance spectrum characteristic domain is calibrated based on the hardware-in-the-loop simulation platform; The calibrated impedance spectrum characteristic domain is input into the grid-connected oscillation risk assessment model, and the predicted oscillation damping ratio of the new energy power station under different control modes is output based on the grid-connected oscillation risk assessment model. An adaptive threshold is constructed based on external operating characteristic parameters that characterize the power system's operating mode. These external operating characteristic parameters include grid topology parameters, power flow distribution parameters, and capacity proportion parameters. The grid connection oscillation risk of the new energy power station is assessed based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold.

2. The method according to claim 1, characterized in that, The process of obtaining operating parameters of different units in a renewable energy power station through perturbation injection and constructing a dynamic equivalent model based on these operating parameters specifically includes: By injecting small disturbance voltage signals of preset amplitude into the ports of different generator units and collecting the corresponding operating parameters, the operating parameters include controller operating parameters, generator terminal voltage operating parameters and generator terminal current operating parameters; Based on the response relationship between the terminal voltage operating parameters and the terminal current operating parameters, the frequency response characteristics of different generator units are constructed. Based on the rate response characteristics and combined with the controller operating parameters, the dynamic equivalent model is constructed.

3. The method according to claim 1, characterized in that, Based on the aforementioned dynamic equivalent model, the impedance spectrum characteristic domain of the new energy power station under different control modes is obtained through frequency domain analysis: In the dynamic equivalent model, corresponding control mode constraints are introduced, including constant power control mode, voltage support control mode and virtual synchronization control mode. Under the constraints of the control mode, the dynamic equivalent model is subjected to small-signal linearization processing to obtain the linearized state equations under different control modes. Based on the linearized state equations under different control modes, the impedance matrix is ​​calculated in the frequency domain, and the corresponding impedance spectrum feature domain is output.

4. The method according to claim 1, characterized in that, Before inputting the calibrated impedance spectrum characteristic domain into the grid-connected oscillation risk assessment model and outputting the predicted oscillation damping ratio of the new energy power station under different control modes based on the grid-connected oscillation risk assessment model, the grid-connected oscillation risk assessment model is constructed, specifically including: Impedance spectrum feature domain samples are extracted based on historical operating data and historical disturbance response data; Obtain historical oscillation event label data corresponding to the impedance spectrum feature domain samples; Based on the impedance spectrum feature domain samples and the historical oscillation event label data, an input feature set and an output label set are constructed, and the model parameters are optimized and trained based on the training algorithm. Regularization constraints and cross-validation mechanisms are introduced during the model parameter optimization training process to construct the grid oscillation risk assessment model.

5. The method according to claim 4, characterized in that, The step of outputting the predicted oscillation damping ratio of the new energy power station under different control modes based on the grid-connected oscillation risk assessment model specifically includes: The calibrated impedance spectrum feature domain is input into the grid oscillation risk assessment model; In the grid oscillation risk assessment model, forward inference is performed based on the trained input feature set and output label set; During the forward inference process, the impedance spectrum feature subsets corresponding to different control modes are extracted and mapped. The predicted value of the oscillation damping ratio under each control mode is calculated based on the mapping operation results.

6. The method according to claim 1, characterized in that, The adaptive threshold construction based on external operating characteristic parameters characterizing the power system's operating mode specifically includes: A network sensitivity matrix characterizing network connectivity and power flow transfer characteristics is established based on the aforementioned power grid topology parameters; Calculate the power margin index of the new energy power station under different operating modes based on the power flow distribution parameters; The impact coefficient of new energy power plants on the overall power grid inertia and damping level is quantified based on the aforementioned capacity ratio parameter. The network sensitivity matrix, the power margin index, and the influence coefficient are weighted and fused to calculate the adaptive threshold.

7. The method according to claim 1, characterized in that, The assessment of the grid connection oscillation risk of the renewable energy power plant based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold specifically includes: Obtain the predicted value of the oscillation damping ratio under different control modes, and compare it with the adaptive threshold one by one; When the predicted value of the oscillation damping ratio is less than the adaptive threshold, it is determined that there is a risk of grid oscillation in the corresponding control mode; When the predicted value of the oscillation damping ratio is greater than or equal to the adaptive threshold, it is determined that there is no grid-connected oscillation risk in the corresponding control mode.

8. A device for assessing the risk of grid connection oscillation at a new energy power station, characterized in that, The device includes an acquisition module and a processing module, wherein, The acquisition module is used to acquire the operating parameters of different units in the new energy power station through perturbation injection, and to construct a dynamic equivalent model based on the operating parameters; The processing module is used to obtain the impedance spectrum characteristic domain of the renewable energy power station under different control modes through frequency domain analysis based on the dynamic equivalent model; construct a hardware-in-the-loop simulation platform and calibrate the impedance spectrum characteristic domain based on the hardware-in-the-loop simulation platform; input the calibrated impedance spectrum characteristic domain into the grid-connected oscillation risk assessment model and output the predicted oscillation damping ratio of the renewable energy power station under different control modes based on the grid-connected oscillation risk assessment model; construct an adaptive threshold based on external operating characteristic parameters characterizing the power system operation mode, the external operating characteristic parameters including grid topology parameters, power flow distribution parameters, and capacity ratio parameters; and assess the grid-connected oscillation risk of the renewable energy power station based on the relationship between the predicted oscillation damping ratio under different control modes and the adaptive threshold.

9. An electronic device, characterized in that, The device includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1 to 7.

Images