Broadband oscillation evaluation method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a broadband oscillation evaluation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring element parameters of each energy storage unit in a target energy storage cluster corresponding to a power system; establishing an energy storage model for the target energy storage cluster according to the element parameters of the energy storage units; performing simulation through the energy storage model, and determining a target broadband oscillation factor; and inputting the target broadband oscillation factor and the energy storage cluster parameters of the different control strategies into a pre-trained mechanism model, and predicting oscillation responses of the power system under the energy storage cluster parameters of the different control strategies through the mechanism model. By adopting the method, the broadband oscillation can be accurately evaluated.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a broadband oscillation evaluation method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] With the increasing penetration of new energy sources in the power system, the structure and characteristics of the power system have undergone significant changes. As an important means to improve the stability and flexibility of the power system, the impact of large-scale energy storage clusters on the power system has attracted much attention. Among these, broadband oscillations seriously threaten the safe and stable operation of the power system, making it crucial to accurately understand the impact mechanism of large-scale energy storage clusters on broadband oscillations in the power system.

[0003] Currently, research on energy storage integration into power systems largely focuses on power regulation and frequency control, with relatively insufficient research into the mechanisms by which energy storage clusters induce or influence broadband oscillations. Existing research methods struggle to comprehensively and accurately analyze the impact of large-scale energy storage clusters on broadband oscillations in power systems under different operating conditions and control strategies. In practical engineering applications, the lack of effective methods for calculating these impact mechanisms makes it impossible to accurately assess the influence of energy storage cluster integration on power system stability, potentially leading to unstable oscillations and affecting the reliability of power supply. Summary of the Invention

[0004] Therefore, it is necessary to provide a broadband oscillation evaluation method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can accurately evaluate broadband oscillations in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a broadband oscillation evaluation method, including:

[0006] Obtain the component parameters of each energy storage unit in the target energy storage cluster corresponding to the power system;

[0007] An energy storage model is established for the target energy storage cluster based on the component parameters of each energy storage unit.

[0008] The target broadband oscillation factor was determined by simulation using the energy storage model.

[0009] The target broadband oscillation factor and the energy storage cluster parameters under different control strategies are input into a pre-trained mechanism model, and the oscillation response of the power system under the energy storage cluster parameters under different control strategies is predicted by the mechanism model.

[0010] In one embodiment, the above-described method of determining the target broadband oscillation factor through simulation using the energy storage model includes:

[0011] Determine the initial broadband oscillation factor; the initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, energy storage unit communication delay, power system load characteristics, and fluctuation characteristics of new energy power generation.

[0012] The initial broadband oscillation factor is adjusted according to preset logic, and the adjusted initial broadband oscillation factor is input into the energy storage model for simulation to obtain the changes of each key node in the power system; the preset logic is to control the change of any one of the initial broadband oscillation factors while keeping the other initial broadband oscillation factors unchanged;

[0013] Based on the changes in each of the key nodes, the target broadband oscillation factor is determined.

[0014] In one embodiment, the training process of the above-mentioned mechanism model includes:

[0015] Obtain historical operating data of the power system;

[0016] The wideband oscillation range is extracted from the historical operating data, and a response label is obtained based on the wideband oscillation range;

[0017] The historical operating data is input into the initial model for training to obtain the predicted response result. The difference between the predicted response result and the response label is calculated, and the parameters of the initial model are adjusted according to the difference until the training is completed to obtain the mechanism model.

[0018] In one embodiment, the extraction of the wideband oscillation range from the historical operating data includes:

[0019] The historical operating data is subjected to a fast Fourier transform to obtain an initial wideband oscillation range, and a first wideband oscillation range is extracted from the initial wideband oscillation range according to a preset filtering range.

[0020] Small-signal stability analysis was performed on the historical operating data to obtain the second wideband oscillation range;

[0021] Impedance scanning analysis is performed on the power system to obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies. Based on the equivalent impedance amplitude and phase curves, the frequency range corresponding to the source-grid impedance amplitude-phase crossover point is identified, and the third broadband oscillation range is determined.

[0022] The wideband oscillation range is obtained based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range.

[0023] In one embodiment, after predicting the oscillation response of the power system under the energy storage cluster parameters of different control strategies using the aforementioned mechanistic model, the method further includes:

[0024] The oscillation response of the power system under the parameters of the energy storage cluster under different control strategies is compared to obtain the comparison results;

[0025] Based on the comparison results, the energy storage cluster parameters corresponding to the target strategy are obtained from the energy storage cluster parameters of different control strategies.

[0026] In one embodiment, the aforementioned oscillation response includes the voltage and current of key nodes in the power system; after predicting the oscillation response of the power system under the energy storage cluster parameters of different control strategies using the mechanistic model, the method further includes:

[0027] Based on the voltage and current of the key nodes of the power system, calculate the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation.

[0028] The stability of the power system is calculated based on the frequency of the broadband oscillation, the amplitude of the broadband oscillation, the system damping ratio, and the power fluctuation coefficient.

[0029] Secondly, this application also provides a broadband oscillation evaluation device, comprising:

[0030] The acquisition module is used to acquire the component parameters of each energy storage unit in the target energy storage cluster corresponding to the power system;

[0031] The model building module is used to build an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit.

[0032] The simulation module is used to perform simulations using the energy storage model to determine the target broadband oscillation factor;

[0033] The estimation module is used to input the target broadband oscillation factor and the energy storage cluster parameters under different control strategies into a pre-trained mechanism model, and predict the oscillation response of the power system under the energy storage cluster parameters under different control strategies through the mechanism model.

[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method in any of the above embodiments.

[0035] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0036] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the methods in any of the above embodiments.

[0037] The aforementioned broadband oscillation assessment method, device, computer equipment, computer-readable storage medium, and computer program product first construct a refined energy storage model for the target energy storage cluster, accurately extract the target oscillation factor within a wide frequency band based on the energy storage model, and finally accurately assess the oscillation response of the power system under different control strategies through a pre-trained excitation model. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart illustrating a broadband oscillation evaluation method in one embodiment;

[0040] Figure 2 This is a structural block diagram of a broadband oscillation evaluation device in one embodiment;

[0041] Figure 3 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0043] In one embodiment, such as Figure 1 As shown, a wideband oscillation evaluation method is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0044] Step 102: Obtain the component parameters of each energy storage unit in the target energy storage cluster corresponding to the power system.

[0045] The target energy storage cluster refers to the sum of a group of energy storage devices connected to the power system to be evaluated, which includes multiple energy storage units.

[0046] Optionally, the target energy storage cluster, as a whole composed of multiple energy storage units, is connected to one or more nodes of the power system via a step-up transformer or grid-connected inverter. Its DC side is connected to the energy storage medium, and its AC side operates synchronously with the power grid, thus becoming a power source-load switchable node in the power system.

[0047] Optionally, the target energy storage cluster includes at least one energy storage unit, which refers to the smallest functional module constituting the energy storage cluster. It can be an energy storage converter unit with a transformer, such as a 2MW / 4MWh rack-mounted battery compartment.

[0048] Furthermore, component parameters are used to describe quantitative indicators of the energy storage unit and its controller, including energy storage type (such as lithium battery, lead-acid battery, etc.), charge and discharge characteristics, power limitation, response time, etc., as well as component parameters of the power system (such as line impedance Z). line =R line +jX line , where R line X is the line resistance. line Line reactance; transformer turns ratio κ, leakage reactance X T ) and operating conditions (such as load changes P) load (t), fluctuation in new energy output P new (t) etc.).

[0049] Optionally, the component parameters of each energy storage unit in the target energy storage cluster can be obtained through the measured data periodically reported by the monitoring platform or database corresponding to the power grid system.

[0050] Step 104: Establish an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit.

[0051] Among them, the energy storage model refers to the equivalent model that corresponds one-to-one with the target energy storage model and can realistically reflect its electrical-control behavior in the simulation environment.

[0052] Optionally, a modular modeling approach can be adopted, treating the energy storage cluster as multiple sub-modules, each corresponding to an energy storage unit. Taking an energy storage unit controlled by a virtual synchronous machine (VSG) as an example, its active power control model is as follows: Where p m p represents virtual mechanical power. e Let J be the electromagnetic power, D be the virtual moment of inertia, ω be the damping coefficient, ω be the output angular frequency, and ω0 be the rated angular frequency. For other components in the power system, such as generators, the Parker model is used.

[0053]

[0054] In the formula, v d v q Let i be the d-axis and q-axis components of the generator terminal voltage. d i q R represents the π, π, d, and q-axis components of the stator current. a L is the stator resistance. d L q ψ is the d-axis and q-axis inductance of the stator winding. d ψ q Let e ​​be the flux linkage along the d and q axes. d e q Let be the d-axis and q-axis electromotive forces of the generator.

[0055] For lithium battery energy storage units, their charge-discharge model can be expressed as follows: Where SOC is the initial state of charge, and I bat For the battery charging and discharging current, C bat This refers to the battery capacity.

[0056] Optionally, after obtaining each submodule, the submodules are connected to a unified bus node in parallel. The connection location and electrical coupling relationship are modeled in combination with the power grid topology to construct an overall model of the target energy storage cluster coupled with the power system, i.e., the energy storage model.

[0057] Step 106: Perform simulation using an energy storage model to determine the target broadband oscillation factor.

[0058] Among them, the target broadband oscillation factor refers to the factors that can cause broadband oscillations in the power system.

[0059] Wideband oscillation refers to the mid-to-high frequency dynamic oscillation phenomenon in a power system within multiple non-power frequency bands (such as 10Hz to 45Hz), caused by factors such as controller dynamic response, grid impedance characteristics, and disturbances from new energy sources. Its characteristics often include: no longer being concentrated at a single frequency point (such as the 1-3Hz low-frequency oscillation in conventional power systems); involving the simultaneous excitation of multiple frequency modes, forming a band-type energy diffusion; and occurring in power grids with high penetration of new energy sources, reduced inertia, and dense energy storage integration. In simpler terms, wideband oscillation means that the system may oscillate within a certain frequency range, rather than at a fixed point.

[0060] The broadband oscillation factor, as a key indicator characterizing the dynamic characteristics of a system, can reflect the oscillation intensity and stability level of a power system under different energy storage configurations and control strategies.

[0061] Optionally, a complete power system-energy storage joint model can be constructed, typical disturbance conditions can be set, time-domain simulation can be performed, and fast Fourier transform and envelope fitting can be performed on the simulation waveform to extract the maximum oscillation amplitude, main oscillation frequency and damping ratio in the target frequency band, thus forming a broadband oscillation factor.

[0062] Step 108: Input the target broadband oscillation factor and the energy storage cluster parameters under different control strategies into the pre-trained mechanism model, and predict the oscillation response of the power system under the energy storage cluster parameters under different control strategies through the mechanism model.

[0063] Among them, the mechanism model refers to the prediction model that has been pre-trained. The mechanism model can reflect the mapping relationship between energy storage control parameters and oscillation response. Different control strategies refer to different combinations of control configurations used for the same energy storage cluster, such as high inertia low damping, low inertia high damping, constant power control, virtual synchronization control, etc.

[0064] Among them, oscillation response refers to the dynamic changes in voltage, current or power exhibited by a power system within a certain frequency range after being subjected to disturbances, such as sudden load changes, power fluctuations, and changes in control parameters. Special attention is paid to whether it exhibits periodic oscillations, whether the oscillations are amplified or attenuated, and what the frequency is.

[0065] Optionally, the target broadband oscillation factor and the energy storage cluster parameters under different control strategies are input into a pre-trained mechanistic model, where the mechanistic model has established a mapping relationship between energy storage control parameters and the broadband oscillation response of the power system. Through model inference, the oscillation response of the power system under different control strategy configurations can be quickly predicted.

[0066] In the above embodiments, a refined energy storage model is first constructed for the target energy storage cluster, and the target oscillation factor in a wide frequency band is accurately extracted based on the energy storage model. Finally, the oscillation response of the power system under different control strategies is accurately obtained through a pre-trained excitation model.

[0067] In one embodiment, the above-mentioned simulation using an energy storage model to determine the target broadband oscillation factor includes: determining an initial broadband oscillation factor; the initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, power system load characteristics, and fluctuation characteristics of new energy power generation; adjusting the initial broadband oscillation factor according to preset logic, and inputting the adjusted initial broadband oscillation factor into the energy storage model for simulation to obtain the changes in each key node in the power system; the preset logic is to control the change of any one initial broadband oscillation factor while keeping other initial broadband oscillation factors unchanged; and determining the target broadband oscillation factor based on the changes in each key node.

[0068] Optionally, a series of key factors affecting the impact of large-scale energy storage clusters on broadband oscillations of the power system can be determined by the operating characteristics of the power system and the working principle of the energy storage cluster, which is the initial broadband oscillation factor in this embodiment.

[0069] Optionally, in addition to the proportion of energy storage capacity, energy storage control strategy parameters (such as virtual inertia coefficient and damping coefficient), and energy storage access location, the initial broadband oscillation factor also considers factors such as the response time of energy storage, the load characteristics of the power system (such as the rate of change of active and reactive power of the load), and the fluctuation characteristics of new energy power generation (such as the rate of change of irradiance of photovoltaic power generation and the rate of change of wind speed of wind power generation).

[0070] Optionally, the preset logic can be the controlled variable method. For example, using the controlled variable method, only one influencing factor is changed each time, while keeping other influencing factors and the operating conditions of the power system constant. A refined model is then used for simulation calculations. For instance, when studying the impact of energy storage capacity ratio on broadband oscillations, the energy storage capacity ratio is gradually increased, starting from a low ratio, such as 5%, and increasing by a certain step each time, such as 2%, until a higher ratio is reached. At each ratio, multiple simulation experiments are conducted, recording the changes in electrical quantities such as voltage and current at key nodes in the power system (such as energy storage access points, load concentration points, etc.).

[0071] Subsequently, the data obtained from each simulation were processed, and the spectral characteristics of the electrical quantities were analyzed using methods such as Fast Fourier Transform (FFT) to obtain parameters such as the frequency and amplitude of the broadband oscillation. The changes in the frequency and amplitude of the broadband oscillation were calculated when each influencing factor changed. For example, the changes in oscillation amplitude ΔA and frequency Δf at a certain key frequency point were calculated when the energy storage capacity ratio increased from 5% to 7%. Using these changes, the sensitivity of each influencing factor to the frequency and amplitude of the broadband oscillation was calculated. For example, the sensitivity of the energy storage capacity ratio to the oscillation amplitude was calculated. Where Δγ represents the change in the proportion of energy storage capacity. By comparing the sensitivity of different influencing factors, the key factor with the most significant impact on broadband oscillation is identified.

[0072] In the above embodiments, an energy storage model closely related to control parameters, access methods and system operating conditions can be constructed in the actual control scenario of energy storage cluster access to the power system. Based on this, a system disturbance simulation of multiple oscillation influencing factors is performed using preset logic, thereby accurately extracting the target broadband oscillation factor that represents the dynamic characteristics of the system.

[0073] In one embodiment, the training process of the above-mentioned mechanism model includes: acquiring historical operating data of the power system; extracting a wideband oscillation range from the historical operating data and obtaining a response label based on the wideband oscillation range; inputting the historical operating data into an initial model for training to obtain a predicted response result; calculating the difference between the predicted response result and the response label; and adjusting the parameters of the initial model according to the difference until training is completed to obtain the mechanism model.

[0074] Historical operating data includes historical operating data of the power system and operating data of energy storage clusters.

[0075] The initial model is a machine learning model that has not yet been trained. It can be a state-space model, a neural network model, or a hybrid model of state-space model and neural network model.

[0076] In this embodiment, the initial model is a hybrid model of state-space model and neural network model. The state-space model is used to describe the linear dynamic characteristics of the system and can accurately reflect the basic physical relationship between the power system and the energy storage cluster; the neural network model is used to capture the nonlinear characteristics and complex interaction relationships in the system.

[0077] Optionally, after obtaining historical operating data, preprocessing can be performed on the data. For example, a large amount of operating data on power systems and energy storage clusters under different operating conditions can be collected, including data on changes in influencing factors and corresponding broadband oscillation response data. Preprocessing of this data, including data cleaning and normalization, can improve the efficiency and accuracy of model training.

[0078] For example, in this embodiment, the original data is first statistically analyzed to calculate its mean of 1.0 pu and standard deviation of 0.05 pu. Based on the principle of three times the standard deviation, the outlier judgment interval is set as [0.85 pu, 1.15 pu]. Data outside this interval is considered outliers, such as the identified data points of 1.3 pu and 0.7 pu. For the detected outliers, linear interpolation replacement is performed using the normal values ​​at adjacent time points. For example, the outlier 1.3 pu is replaced with the mean of 1.125 pu, which is the sum of 1.12 pu at the previous time point and 1.13 pu at the next time point, thereby achieving effective repair and smoothing of the original data.

[0079] For example, in this embodiment, to reduce high-frequency noise interference in the data, a 5-point moving average filtering method is used to smooth the original data sequence. Specifically, by averaging each data point and its two nearest neighbors, local fluctuations are eliminated and data stability is improved. For example, for the original data sequence [0.98, 1.02, 1.01, 0.99, 1.03], the smoothed sequence obtained after filtering by this method is [1.002, 1.004, 1.006, 1.008, 1.01].

[0080] In this embodiment, to unify data units and improve the stability of model training, the cleaned raw data is normalized and mapped to the [0,1] interval. The min-max normalization method is used, and the calculation formula is as follows: Where x represents the original data, x min and x max They are respectively in the data

[0081] Minimum and maximum values. Taking a minimum value of 0.85 pu and a maximum value of 1.15 pu as an example, if the original value is 1.15 pu, the normalization result is 1.0; if the original value is 0.9 pu, the normalization result is approximately 0.167; and if the original value is 1.0 pu, the corresponding normalization value is 0.5. This method can achieve data standardization, facilitating subsequent unified modeling and processing.

[0082] Historical operating data accurately reflects the dynamic response of the power system under different control strategies, load levels, and energy storage conditions. Frequency domain analysis can extract key oscillation-related features from these historical signals, such as the dominant oscillation frequency, oscillation amplitude, and damping ratio. These features are essentially the system's true output response under specific input conditions and can serve as labels for model training in supervised learning.

[0083] Next, the hybrid model is trained using the preprocessed data. Parameter estimation methods for state-space models, such as least squares, are used to initially estimate the parameters of the state-space model. Then, the output of the state-space model is used as the input to the neural network model, and the weights and thresholds of the neural network model are trained and optimized using the backpropagation algorithm. During training, cross-validation is employed, dividing the dataset into training, validation, and test sets, and continuously adjusting the model parameters to optimize the model's performance on the validation set, thus avoiding overfitting.

[0084] Next, the difference between the model's prediction result (i.e., the predicted response result in this embodiment) and the response label is used to evaluate the model's accuracy by calculating metrics such as root mean square error (RMSE) and mean absolute error (MAE). For example, the formula for calculating the root mean square error is... Where y i This is the actual value. is the model's predicted value, and n is the number of test data. If the model's error index is within an acceptable range, it indicates that the model can well reflect the impact mechanism of large-scale energy storage clusters on the broadband oscillation of the power system; if the error is large, it is necessary to readjust the model structure or parameters and retrain and validate it.

[0085] In the above embodiments, broadband oscillation features can be automatically extracted based on historical operating data, and a high-precision mechanism model can be constructed by combining supervised training methods. This enables the rapid and accurate prediction of the broadband oscillation response of the power system under different control strategy parameter inputs, thereby improving the efficiency of strategy evaluation, reducing simulation costs, and possessing good engineering adaptability and generalization ability.

[0086] In one embodiment, the extraction of the broadband oscillation range includes: performing a fast Fourier transform on the signal to obtain an initial broadband oscillation range, and extracting a first broadband oscillation range from the initial broadband oscillation range according to a preset screening range; performing small-signal stability analysis on the signal to obtain a second broadband oscillation range; performing impedance scanning analysis on the power system to obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies, identifying the frequency range corresponding to the source-grid impedance amplitude-phase crossover point based on the equivalent impedance amplitude and phase curves, and determining a third broadband oscillation range; and obtaining the broadband oscillation range based on the first broadband oscillation range, the second broadband oscillation range, and the third broadband oscillation range.

[0087] In this embodiment, a multi-dimensional extraction method is used to extract the wideband oscillation range from historical operating data.

[0088] First, a fast Fourier transform is performed on the historical operating data to conduct a spectral analysis, obtaining the frequency distribution of the power system at different times. Then, by using a preset threshold, the frequency range with a higher power spectral density is selected as the first broadband oscillation range.

[0089] Secondly, to analyze the oscillation stability of the power system under different control strategies, a small-signal stability analysis method is used to linearize the nonlinear dynamic system, which includes energy storage units, a grid model, and a controller, near the operating point, resulting in a set of linear equations in standard state-space form. Furthermore, characteristic equations are constructed to solve for the eigenvalues ​​of the system matrix. The stability of the system under disturbances is determined based on the real parts of the eigenvalues, while the oscillation frequencies corresponding to each mode are calculated from the imaginary parts. Finally, the frequency range and stability of the power system that may exhibit wideband oscillations are determined, yielding the second wideband oscillation range.

[0090] For example, the power system model is linearized. Original model: The power system model contains nonlinear equations such as the generator Park equation and the energy storage VSG control equation (e.g., the voltage and flux linkage equations in the Park model contain nonlinear terms). Linearization process: Near the system equilibrium point (e.g., the rated operating point), the nonlinear equations are expanded into Taylor series, ignoring higher-order terms, resulting in the linearized state equation: x' = Ax + Bu. Where x is the state variable (e.g., generator power angle, voltage amplitude, energy storage state of charge, etc.); A is the system matrix (composed of the linearized model parameters); B is the input matrix; and u is the external disturbance (e.g., load changes, fluctuations in renewable energy output). Then, the characteristic equation is constructed and the eigenvalues ​​are solved. The stability of the linearized state equation is determined by the characteristic equation: |sI-A| = 0, where s is a complex variable and I is the identity matrix. Solving this equation yields the eigenvalue λ. i =σ i +jω i (σ i For the real part, ω i (The imaginary part). The stability of the system is determined based on the real part of the eigenvalues. A positive real part indicates that the system is unstable and may oscillate. The imaginary part corresponds to the angular frequency of the oscillation, and the corresponding second wideband oscillation range can be obtained by calculation.

[0091] Furthermore, in this embodiment, to identify the broadband oscillation risk caused by source-grid interaction, frequency scanning analysis is performed on the energy storage access node in the power system based on impedance stability theory to obtain the equivalent impedance amplitude and phase response curves of the node at different frequencies. Further analysis of the cross-impedance characteristics between the source and grid ends identifies frequency points with equal amplitudes and phases close to 180°, and based on this, delineates unstable frequency bands that may lead to oscillations, thereby determining the third broadband oscillation range. This method can effectively reflect the dynamic stability risk caused by source-grid impedance coupling.

[0092] Finally, by comprehensively analyzing the results obtained from the above methods, the wideband oscillation frequency range of the power system before and after the access of the energy storage cluster was determined.

[0093] Alternatively, in other embodiments, a fusion calculation can be performed based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range, taking into account the frequency domain energy distribution, small signal stability results, and impedance characteristic changes to obtain a more accurate wideband oscillation range.

[0094] Furthermore, based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range, the intersection of the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range is taken as the final wideband oscillation range; if the intersection is empty, the union is taken and combined with engineering experience to screen key frequency intervals to obtain the wideband oscillation range.

[0095] In the above embodiments, by introducing three complementary methods—Fast Fourier Transform, Small Signal Stability Analysis, and Impedance Scanning—a wideband oscillation range is extracted from multiple dimensions, including frequency domain energy distribution, system modal characteristics, and source-network impedance coupling, achieving multi-source fusion and enhanced verification for oscillation frequency identification. Specifically, Fast Fourier Transform can quickly identify high-energy frequency bands, Small Signal Analysis accurately reveals potential modal instability frequencies, and the Impedance Scanning method reflects the dynamic stability risks brought about by source-network interaction. By integrating these three types of wideband oscillation ranges, the accuracy of the final identification result is significantly improved.

[0096] In one embodiment, after predicting the oscillation response of the power system under different control strategies using a mechanistic model, the method further includes: comparing the oscillation response of the power system under different control strategies using energy storage cluster parameters to obtain a comparison result; and obtaining the energy storage cluster parameters corresponding to the target strategy from the energy storage cluster parameters of different control strategies based on the comparison result.

[0097] In this embodiment, the oscillation response refers to the dynamic change of electrical quantities such as voltage and current at key nodes of the power system over time under disturbance. The voltage change curves, current swing, steady-state recovery time, and oscillation frequency offset of the key nodes under each control strategy are compared horizontally. Based on preset targets, such as minimum voltage offset, shortest recovery time, or lowest oscillation intensity, the optimal target control strategy is determined, and its corresponding energy storage cluster parameters are obtained accordingly.

[0098] Furthermore, the target broadband oscillation factor, parameters of energy storage clusters of different sizes and control strategies are input into the constructed influence mechanism model for prediction. The changes in electrical quantities such as voltage and current at key nodes of the power system output by the model are analyzed, which is the oscillation response in the above embodiment. Characteristic parameters such as the frequency and amplitude of broadband oscillations are extracted. The changes in broadband oscillation characteristic parameters before and after the access of energy storage clusters in different scenarios are compared to evaluate the suppression or exacerbation effect of energy storage clusters on broadband oscillations, which is the comparison result in this embodiment. If the oscillation amplitude of key nodes is significantly reduced after the access of energy storage clusters, and the oscillation frequency stabilizes within a safe range, it indicates that the energy storage cluster has a good suppression effect on broadband oscillations in this scenario. The comparison before and after the access of energy storage clusters is achieved through two runs of the same model under different parameter configurations: one without energy storage parameters (baseline), and one with energy storage parameters (after access). This comparison method can accurately quantify the degree of influence of energy storage clusters on broadband oscillations and is the core method for evaluating their stability effect. If it is found that an energy storage cluster under a certain control strategy can effectively suppress broadband oscillations in a specific scenario, it is used as the energy storage cluster parameter corresponding to the target strategy.

[0099] In the above embodiments, by comparing the oscillation response of the energy storage cluster under different control strategies, the energy storage cluster parameters corresponding to the target strategy can be accurately determined.

[0100] In one embodiment, the aforementioned oscillation response includes the voltage and current of key nodes in the power system; after predicting the oscillation response of the power system under energy storage cluster parameters of different control strategies through a mechanistic model, the method further includes: calculating the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation based on the voltage and current of the key nodes in the power system; and calculating the stability of the power system based on the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation.

[0101] In this embodiment, by conducting in-depth analysis of the timing data of key node voltages and currents predicted by the mechanistic model, key dynamic characteristic parameters of the power system under broadband disturbances are extracted.

[0102] For example, firstly, a fast Fourier transform is performed on the voltage and current waveforms to identify the frequency components with concentrated energy in the spectrum and determine the frequency range in which the system may oscillate; then, the peak and valley values ​​of the oscillations are extracted based on the time-domain waveforms, and the oscillation amplitude is calculated to measure the intensity of the fluctuations; further, the attenuation envelope curve of the waveform is fitted, the damping ratio is calculated, and the system's ability to attenuate oscillations is evaluated; at the same time, a volatility analysis is performed on the active power curve, and the power fluctuation coefficient is calculated to reflect the stability of the system operation.

[0103] Optionally, the average value can be calculated based on the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation to obtain the index value. Then, the index value is compared with a preset threshold to obtain the stability of the circuit system.

[0104] Optionally, by normalizing the broadband oscillation frequency, oscillation amplitude, system damping ratio, and power fluctuation coefficient to a score of 0–1, frequency score, amplitude score, damping score, and fluctuation score are obtained respectively; then, a comprehensive stability index is generated according to preset weights. The comprehensive stability index is then compared with a preset threshold unit to obtain the stability result.

[0105] Among them, the preset weights can be preset based on engineering standards; the stability results can be stability levels, which include three types: stable, critically stable, and unstable.

[0106] For example, if S ≥ 0.8, it is considered stable; 0.6 ≤ S < 0.8 is considered critically stable; and S < 0.6 is considered unstable. This multi-index weighted scoring method can integrate dimensions such as oscillation intensity, attenuation capability, and power fluctuation into a single value, which facilitates rapid quantitative comparison and optimization decision-making of system stability under different energy storage control strategies.

[0107] In the above embodiments, the stability of the power system can be accurately calculated by using the frequency of the broadband oscillation, the amplitude of the broadband oscillation, the system damping ratio, and the power fluctuation coefficient.

[0108] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0109] Based on the same inventive concept, this application also provides a broadband oscillation evaluation apparatus for implementing the broadband oscillation evaluation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more broadband oscillation evaluation apparatus embodiments provided below can be found in the limitations of the broadband oscillation evaluation method described above, and will not be repeated here.

[0110] In one exemplary embodiment, such as Figure 2 As shown, a broadband oscillation evaluation device is provided, including: an acquisition module 100, a model building module 200, a simulation module 300, and an estimation module 400, wherein:

[0111] The acquisition module is used to acquire the component parameters of each energy storage unit in the target energy storage cluster corresponding to the power system.

[0112] The model building module is used to build an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit.

[0113] The simulation module is used to perform simulations using an energy storage model to determine the target broadband oscillation factor.

[0114] The estimation module is used to input the target broadband oscillation factor and the energy storage cluster parameters under different control strategies into a pre-trained mechanistic model, and predict the oscillation response of the power system under the energy storage cluster parameters under different control strategies through the mechanistic model.

[0115] In one embodiment, the simulation module includes:

[0116] The initial factor determination unit is used to determine the initial broadband oscillation factor. The initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, power system load characteristics, and fluctuation characteristics of new energy power generation.

[0117] The factor simulation unit is used to adjust the initial broadband oscillation factor according to preset logic, and input the adjusted initial broadband oscillation factor into the energy storage model for simulation to obtain the changes of each key node in the power system; the preset logic is to control the change of any one initial broadband oscillation factor while keeping the other initial broadband oscillation factors unchanged.

[0118] The comparison unit is used to determine the target broadband oscillation factor based on the changes in each key node.

[0119] In one embodiment, the above-described apparatus further includes a training module, which includes:

[0120] The sample acquisition unit is used to acquire historical operating data of the power system.

[0121] The tag determination unit is used to extract the wideband oscillation range from historical operating data and obtain the response tag based on the wideband oscillation range.

[0122] The model training unit is used to input historical running data into the initial model for training, obtain the predicted response result, calculate the difference between the predicted response result and the response label, and adjust the parameters of the initial model according to the difference until the training is completed and the mechanism model is obtained.

[0123] In one embodiment, the label determination unit includes:

[0124] The first oscillation range determination subunit is used to perform a fast Fourier transform on historical operating data to obtain an initial wideband oscillation range, and extract the first wideband oscillation range from the initial wideband oscillation range according to a preset screening range.

[0125] The second oscillation range determination subunit is used to perform small-signal stability analysis on historical operating data to obtain the second wideband oscillation range.

[0126] The third oscillation range determination subunit is used to perform impedance scanning analysis on the power system, obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies, identify the frequency range corresponding to the source-grid impedance amplitude-phase crossover point based on the equivalent impedance amplitude and phase curves, and determine the third wideband oscillation range.

[0127] The synthesis subunit is used to obtain the wideband oscillation range based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range.

[0128] In one embodiment, the above-described apparatus further includes a parameter determination module, which includes:

[0129] The response comparison unit is used to compare the oscillation response of the power system under the parameters of the energy storage cluster under different control strategies, and obtain the comparison results.

[0130] The parameter comparison unit is used to obtain the energy storage cluster parameters corresponding to the target strategy from the energy storage cluster parameters of different control strategies based on the comparison results.

[0131] In one embodiment, the aforementioned oscillation response includes the voltage and current of critical nodes in the power system; the aforementioned device further includes a stability analysis module, which includes:

[0132] The calculation unit is used to calculate the frequency, amplitude, system damping ratio, and power fluctuation coefficient of broadband oscillation based on the voltage and current of key nodes in the power system.

[0133] The stability calculation unit is used to calculate the stability of the power system based on the frequency, amplitude, damping ratio, and power fluctuation coefficient of the broadband oscillation.

[0134] Each module in the aforementioned broadband oscillation evaluation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0135] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores historical runtime data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a broadband oscillation evaluation method.

[0136] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0137] In one exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: acquiring the component parameters of each energy storage unit in a target energy storage cluster corresponding to a power system; establishing an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit; performing simulation through the energy storage model to determine the target broadband oscillation factor; inputting the target broadband oscillation factor and energy storage cluster parameters under different control strategies into a pre-trained mechanistic model, and predicting the oscillation response of the power system under the energy storage cluster parameters of different control strategies through the mechanistic model.

[0138] In one embodiment, when the processor executes the computer program, it further performs the following steps: determining an initial broadband oscillation factor; the initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, power system load characteristics, and fluctuation characteristics of new energy power generation; adjusting the initial broadband oscillation factor according to preset logic, and inputting the adjusted initial broadband oscillation factor into the energy storage model for simulation to obtain the changes of each key node in the power system; the preset logic is to control the change of any one initial broadband oscillation factor while keeping other initial broadband oscillation factors unchanged; and determining a target broadband oscillation factor based on the changes of each key node.

[0139] In one embodiment, when the processor executes the computer program, it also performs the following steps: acquiring historical operating data of the power system; extracting a wideband oscillation range from the historical operating data and obtaining a response label based on the wideband oscillation range; inputting the historical operating data into an initial model for training to obtain a predicted response result; calculating the difference between the predicted response result and the response label; and adjusting the parameters of the initial model according to the difference until training is completed to obtain a mechanistic model.

[0140] In one embodiment, when the processor executes the computer program, it further performs the following steps: performing a fast Fourier transform on historical operating data to obtain an initial wideband oscillation range, and extracting a first wideband oscillation range from the initial wideband oscillation range according to a preset filtering range; performing small-signal stability analysis on historical operating data to obtain a second wideband oscillation range; performing impedance scanning analysis on the power system to obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies, identifying the frequency range corresponding to the source-grid impedance amplitude-phase crossover point based on the equivalent impedance amplitude and phase curves, and determining a third wideband oscillation range; and obtaining the wideband oscillation range based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range.

[0141] In one embodiment, when the processor executes the computer program, it further performs the following steps: comparing the oscillation response of the power system under the energy storage cluster parameters of different control strategies to obtain the comparison results; and based on the comparison results, obtaining the energy storage cluster parameters corresponding to the target strategy from the energy storage cluster parameters of different control strategies.

[0142] In one embodiment, when the processor executes the computer program, it further performs the following steps: calculating the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation based on the voltage and current of the key nodes of the power system; and calculating the stability of the power system based on the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation.

[0143] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps: obtaining the component parameters of each energy storage unit in a target energy storage cluster corresponding to a power system; establishing an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit; performing simulation through the energy storage model to determine the target broadband oscillation factor; inputting the target broadband oscillation factor and energy storage cluster parameters under different control strategies into a pre-trained mechanistic model, and predicting the oscillation response of the power system under the energy storage cluster parameters of different control strategies through the mechanistic model.

[0144] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining an initial broadband oscillation factor; the initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, power system load characteristics, and fluctuation characteristics of new energy power generation; adjusting the initial broadband oscillation factor according to preset logic, and inputting the adjusted initial broadband oscillation factor into the energy storage model for simulation to obtain the changes of each key node in the power system; the preset logic is to control the change of any one initial broadband oscillation factor while keeping other initial broadband oscillation factors unchanged; and determining a target broadband oscillation factor based on the changes of each key node.

[0145] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: acquiring historical operating data of the power system; extracting a wideband oscillation range from the historical operating data and obtaining a response label based on the wideband oscillation range; inputting the historical operating data into an initial model for training to obtain a predicted response result; calculating the difference between the predicted response result and the response label; and adjusting the parameters of the initial model according to the difference until training is completed to obtain a mechanistic model.

[0146] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: performing a fast Fourier transform on historical operating data to obtain an initial wideband oscillation range, and extracting a first wideband oscillation range from the initial wideband oscillation range according to a preset screening range; performing small-signal stability analysis on historical operating data to obtain a second wideband oscillation range; performing impedance scanning analysis on the power system to obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies, identifying the frequency range corresponding to the source-grid impedance amplitude-phase crossover point based on the equivalent impedance amplitude and phase curves, and determining a third wideband oscillation range; and obtaining the wideband oscillation range based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range.

[0147] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: comparing the oscillation response of the power system under the energy storage cluster parameters of different control strategies to obtain the comparison results; and based on the comparison results, obtaining the energy storage cluster parameters corresponding to the target strategy from the energy storage cluster parameters of different control strategies.

[0148] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: calculating the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation based on the voltage and current of the key nodes of the power system; and calculating the stability of the power system based on the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation.

[0149] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps: acquiring the component parameters of each energy storage unit in a target energy storage cluster corresponding to a power system; establishing an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit; performing simulation through the energy storage model to determine the target broadband oscillation factor; inputting the target broadband oscillation factor and energy storage cluster parameters under different control strategies into a pre-trained mechanistic model, and predicting the oscillation response of the power system under energy storage cluster parameters under different control strategies through the mechanistic model.

[0150] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining an initial broadband oscillation factor; the initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, power system load characteristics, and fluctuation characteristics of new energy power generation; adjusting the initial broadband oscillation factor according to preset logic, and inputting the adjusted initial broadband oscillation factor into the energy storage model for simulation to obtain the changes of each key node in the power system; the preset logic is to control the change of any one initial broadband oscillation factor while keeping other initial broadband oscillation factors unchanged; and determining a target broadband oscillation factor based on the changes of each key node.

[0151] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: acquiring historical operating data of the power system; extracting a wideband oscillation range from the historical operating data and obtaining a response label based on the wideband oscillation range; inputting the historical operating data into an initial model for training to obtain a predicted response result; calculating the difference between the predicted response result and the response label; and adjusting the parameters of the initial model according to the difference until training is completed to obtain a mechanistic model.

[0152] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: performing a fast Fourier transform on historical operating data to obtain an initial wideband oscillation range, and extracting a first wideband oscillation range from the initial wideband oscillation range according to a preset screening range; performing small-signal stability analysis on historical operating data to obtain a second wideband oscillation range; performing impedance scanning analysis on the power system to obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies, identifying the frequency range corresponding to the source-grid impedance amplitude-phase crossover point based on the equivalent impedance amplitude and phase curves, and determining a third wideband oscillation range; and obtaining the wideband oscillation range based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range.

[0153] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: comparing the oscillation response of the power system under the energy storage cluster parameters of different control strategies to obtain the comparison results; and based on the comparison results, obtaining the energy storage cluster parameters corresponding to the target strategy from the energy storage cluster parameters of different control strategies.

[0154] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: calculating the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation based on the voltage and current of the key nodes of the power system; and calculating the stability of the power system based on the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation.

[0155] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0156] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0157] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A broadband oscillation evaluation method, characterized in that, The method includes: Obtain the component parameters of each energy storage unit in the target energy storage cluster corresponding to the power system; An energy storage model is established for the target energy storage cluster based on the component parameters of each energy storage unit. The target broadband oscillation factor was determined by simulation using the energy storage model. The target broadband oscillation factor and the energy storage cluster parameters under different control strategies are input into a pre-trained mechanism model, and the oscillation response of the power system under the energy storage cluster parameters under different control strategies is predicted by the mechanism model.

2. The method according to claim 1, characterized in that, The step of determining the target broadband oscillation factor through simulation using the energy storage model includes: Determine the initial broadband oscillation factor; the initial broadband oscillation factor includes one or more of the following: energy storage capacity ratio, energy storage control strategy parameters, energy storage access location, energy storage response time, power system load characteristics, and fluctuation characteristics of new energy power generation. The initial broadband oscillation factor is adjusted according to preset logic, and the adjusted initial broadband oscillation factor is input into the energy storage model for simulation to obtain the changes of each key node in the power system; the preset logic is to control the change of any one of the initial broadband oscillation factors while keeping the other initial broadband oscillation factors unchanged; Based on the changes in each of the key nodes, the target broadband oscillation factor is determined.

3. The method according to claim 1, characterized in that, The training process of the mechanistic model includes: Obtain historical operating data of the power system; The wideband oscillation range is extracted from the historical operating data, and a response label is obtained based on the wideband oscillation range; The historical operating data is input into the initial model for training to obtain the predicted response result. The difference between the predicted response result and the response label is calculated, and the parameters of the initial model are adjusted according to the difference until the training is completed to obtain the mechanism model.

4. The method according to claim 3, characterized in that, The extraction of the wideband oscillation range from the historical operating data includes: The historical operating data is subjected to a fast Fourier transform to obtain an initial wideband oscillation range, and a first wideband oscillation range is extracted from the initial wideband oscillation range according to a preset filtering range. Small-signal stability analysis was performed on the historical operating data to obtain the second wideband oscillation range; Impedance scanning analysis is performed on the power system to obtain the equivalent impedance amplitude and phase curves of the energy storage access node at different frequencies. Based on the equivalent impedance amplitude and phase curves, the frequency range corresponding to the source-grid impedance amplitude-phase crossover point is identified, and the third broadband oscillation range is determined. Based on the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range, the intersection of the first wideband oscillation range, the second wideband oscillation range, and the third wideband oscillation range is taken as the final wideband oscillation range; if the intersection is empty, the union is taken and combined with engineering experience to screen key frequency intervals to obtain the wideband oscillation range.

5. The method according to claim 1, characterized in that, After predicting the oscillation response of the power system under the parameters of the energy storage cluster under different control strategies using the aforementioned mechanism model, the method further includes: The oscillation response of the power system under the parameters of the energy storage cluster under different control strategies is compared to obtain the comparison results; Based on the comparison results, the energy storage cluster parameters corresponding to the target strategy are obtained from the energy storage cluster parameters of different control strategies.

6. The method according to claim 1, characterized in that, The oscillation response includes the voltage and current of key nodes in the power system; after predicting the oscillation response of the power system under the energy storage cluster parameters of different control strategies using the mechanistic model, it further includes: Based on the voltage and current of the key nodes of the power system, calculate the frequency, amplitude, system damping ratio, and power fluctuation coefficient of the broadband oscillation. Based on the frequency of the broadband oscillation, the amplitude of the broadband oscillation, the system damping ratio, and the power fluctuation coefficient, a comprehensive stability index is obtained using a weighted summation algorithm. This index is then compared with a preset threshold to determine the stability of the power system.

7. A broadband oscillation evaluation device, characterized in that, The device includes: The acquisition module is used to acquire the component parameters of each energy storage unit in the target energy storage cluster corresponding to the power system; The model building module is used to build an energy storage model for the target energy storage cluster based on the component parameters of each energy storage unit. The simulation module is used to perform simulations using the energy storage model to determine the target broadband oscillation factor; The estimation module is used to input the target broadband oscillation factor and the energy storage cluster parameters under different control strategies into a pre-trained mechanism model, and predict the oscillation response of the power system under the energy storage cluster parameters under different control strategies through the mechanism model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images