Wind power prediction assisted wind power grid connection management method and device
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INNER MONGOLIA UNIV OF TECH
- Filing Date
- 2026-03-25
- Publication Date
- 2026-06-02
Smart Images
Figure CN121965749B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of power grid technology, and in particular to a wind power grid connection management method and device with wind power forecasting assistance. Background Technology
[0002] With the continuous growth of wind power installed capacity and the increasing proportion of new energy grid connection, wind power, as an important renewable energy source, plays an increasingly prominent role in the power system, and large-scale wind power grid connection has become an important part of modern power grid operation.
[0003] Currently, existing wind power grid connection management methods rely on dispatchers scheduling grid connection times based on wind power forecasts or operational experience. Some systems make rough judgments about grid connection timing using simple power forecasts, but most only focus on the output changes of a single wind farm, lacking systematic analysis of the statistical patterns of wind power forecast errors, and also lacking technical mechanisms to quantitatively assess grid connection risks using the probabilistic characteristics of forecast errors. In actual operation, there is often a certain deviation between wind power forecast results and actual power generation. The forecast error exhibits obvious random fluctuations under different time periods and operating conditions. When grid connection dispatch relies solely on single forecast results or simple threshold judgments for grid connection control, it is difficult to accurately reflect the potential risks brought about by the uncertainty of wind power output.
[0004] In summary, the existing technology suffers from a lack of a grid connection risk quantification assessment mechanism based on the probability distribution of wind power prediction errors. This makes it difficult to accurately reflect the potential impact of wind power output uncertainty on grid operation stability during wind power grid connection scheduling, further affecting the rationality of wind power grid connection timing and grid operation safety. Summary of the Invention
[0005] The purpose of this application is to provide a wind power prediction-assisted wind power grid connection management method and device to solve the technical problem in the prior art that the lack of a grid connection risk quantification assessment mechanism based on the probability distribution of wind power prediction error makes it difficult to accurately reflect the potential impact of wind power output uncertainty on grid operation stability during wind power grid connection scheduling, which further affects the rationality of wind power grid connection timing and grid operation safety.
[0006] In view of the above problems, this application provides a wind power grid connection management method and device with wind power prediction assistance.
[0007] In a first aspect, this application provides a wind power prediction-assisted grid connection management method, implemented through a wind power prediction-assisted grid connection management device, comprising: acquiring historical wind power prediction error sequences of multiple wind power grids within a power grid area; constructing multiple prediction error probability distribution models corresponding to the multiple wind power grids based on the historical wind power prediction error sequences; calculating multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids based on the multiple prediction error probability distribution models; generating a grid connection access timing window for each wind power grid based on the multiple sets of grid connection risk probability vectors; performing timing window overlap analysis on the multiple grid connection access timing windows corresponding to the multiple wind power grids to obtain a set of wind power grid access pairs that meet the grid connection access conditions; generating a grid connection control command based on the set of wind power grid access pairs; and managing the grid connection access timing of each wind power grid using the grid connection control command.
[0008] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: wherein the historical wind power prediction error sequence is obtained by calculating the error between the historical actual wind power data and the historical wind power prediction power data of each wind power grid at the corresponding time point; the prediction error feature vector of the historical wind power prediction error sequence is extracted; the prediction error sample set is estimated according to the prediction error feature vector to construct a prediction error probability density function, and a prediction error probability distribution model of each wind power grid is constructed according to the prediction error probability density function.
[0009] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: calculating multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids based on the multiple prediction error probability distribution models, wherein each set of grid connection risk probability vectors includes the probability of power exceeding the limit, the probability of power surge, and the probability of power drop when any wind power grid is connected to the grid.
[0010] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: obtaining the grid connection node capacity and grid connection line margin; constructing a grid connection impact index assessment model based on the grid connection node capacity and grid connection line margin; inputting multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids into the grid connection impact index assessment model to calculate the grid connection impact index of each wind power grid; adaptively adjusting the grid connection access window width based on the grid connection impact index of each wind power grid to obtain an adaptive grid connection access window width; and generating a wind power grid connection access timing window based on the adaptive grid connection access window width of each wind power grid.
[0011] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: analyzing the stable moment of each wind power grid, wherein the stable moment is the moment when the predicted power change rate is less than a preset change rate threshold; and generating a wind power grid connection timing window with the stable moment of each wind power grid as the window center and the adaptive grid connection access window width of each wind power grid.
[0012] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: aligning the multiple grid connection access time-series windows corresponding to the multiple wind power grids to obtain a wind power grid connection access window set; constructing a wind power grid time-series window topology map based on the wind power grid connection access window set; performing grid connection access conflict analysis on the wind power grid time-series window topology map to obtain a wind power grid connection access conflict matrix; and identifying a set of wind power grid access pairs that meet the grid connection access conditions based on the wind power grid connection access conflict matrix.
[0013] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: constructing a wind power grid time-series window topology graph based on the wind power grid grid connection access window set, wherein the wind power grid time-series window topology graph includes topology nodes and topology connection edges; the topology nodes are the wind power grid grid connection access windows in the wind power grid grid connection access window set, and the topology connection edges are established by determining whether the grid connection access time-series windows of two wind power grids in the wind power grid grid connection access window set overlap.
[0014] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: traversing all candidate node pairs in the grid connection conflict matrix; when the grid connection windows of the two wind power grids of any candidate node pair overlap in time; determining whether the total capacity of the grid-connected nodes and the total margin of the grid-connected lines are both less than the corresponding preset thresholds; if both are less than the corresponding preset thresholds, marking the current candidate node pair as a wind power grid access pair, and outputting a set of wind power grid access pairs.
[0015] Preferably, the wind power prediction-assisted wind power grid connection management method further includes: acquiring multiple historical wind power prediction error sequences for each wind power grid within the power grid area at multiple time scales; performing multi-scale optimization of the prediction error probability distribution model corresponding to each wind power grid based on the multiple historical wind power prediction error sequences at multiple time scales; wherein, by selecting a time scale, multiple sets of grid connection risk probability vectors output by the prediction error probability distribution model are obtained at the corresponding time scale, and a grid connection access timing window for each wind power grid at the corresponding time scale is generated based on the multiple sets of grid connection risk probability vectors.
[0016] Secondly, this application also provides a wind power prediction-assisted grid connection management device for executing the wind power prediction-assisted grid connection management method as described in the first aspect, comprising: a prediction error probability distribution model construction module, used to acquire historical wind power prediction error sequences of multiple wind power grids within a power grid area, and construct multiple prediction error probability distribution models corresponding to the multiple wind power grids based on the historical wind power prediction error sequences; a grid connection risk probability vector calculation module, used to calculate multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids according to the multiple prediction error probability distribution models; a wind power grid access pair set acquisition module, used to generate a grid connection access time sequence window for each wind power grid based on the multiple sets of grid connection risk probability vectors, perform time sequence window overlap analysis on the multiple grid connection access time sequence windows corresponding to the multiple wind power grids, and obtain a set of wind power grid access pairs that meet the grid connection access conditions; and a management module, used to generate grid connection control instructions based on the set of wind power grid access pairs, and manage the grid connection access time sequence of each wind power grid with the grid connection control instructions.
[0017] The technical solution provided in this application has at least the following technical effects or advantages: by achieving the technical goal of probabilistically modeling the uncertainty of wind power prediction error and quantitatively assessing the grid connection risk of wind power grid, it achieves the technical effect of improving the scientific nature of wind power grid connection timing planning and enhancing the grid's operational stability and scheduling coordination capabilities in multi-wind power grid connection scenarios.
[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description
[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0020] Figure 1 This is a flowchart illustrating the wind power grid connection management method for wind power prediction assistance in this application.
[0021] Figure 2 This is a schematic diagram of the wind power grid connection control device for wind power prediction assistance in this application.
[0022] Figure labeling: Prediction error probability distribution model construction module 1, grid connection risk probability vector calculation module 2, wind power grid access pair set acquisition module 3, control module 4. Detailed Implementation
[0023] This application provides a wind power prediction-assisted wind power grid connection management method and device, solving the technical problem in existing technologies where the lack of a quantitative assessment mechanism for grid connection risk based on the probability distribution of wind power prediction errors makes it difficult to accurately reflect the potential impact of wind power output uncertainty on grid operation stability during wind power grid connection scheduling, further affecting the rationality of wind power grid connection timing and grid operation safety. The application achieves the technical objective of probabilistically modeling the uncertainty of wind power prediction errors and quantitatively assessing the grid connection risk of wind power grids, thereby improving the scientific nature of wind power grid connection timing planning and enhancing the grid's operational stability and scheduling coordination capabilities in multi-wind power grid connection scenarios.
[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.
[0025] Example 1, please refer to the appendix. Figure 1 This application provides a wind power prediction-assisted grid connection management method, which is applied to a wind power prediction-assisted grid connection management device, and specifically includes the following steps:
[0026] Historical wind power prediction error sequences of multiple wind power grids within the power grid area are obtained, and multiple prediction error probability distribution models corresponding to the multiple wind power grids are constructed based on the historical wind power prediction error sequences.
[0027] Furthermore, this application also includes: wherein the historical wind power prediction error sequence is obtained by calculating the error between the historical actual wind power data and the historical wind power prediction data of each wind power grid at the corresponding time point; the prediction error feature vector of the historical wind power prediction error sequence is extracted; the prediction error sample set is estimated according to the prediction error feature vector to construct a prediction error probability density function, and a prediction error probability distribution model of each wind power grid is constructed according to the prediction error probability density function.
[0028] Specifically, historical wind power prediction error sequences of multiple wind power grids within the power grid area are obtained. Based on these historical wind power prediction error sequences, a probability distribution model of the prediction error for each wind power grid is constructed. This represents the process of collecting and organizing historical operating data of each wind power grid within the power grid area in the power grid dispatch and operation monitoring system. Error time series data is formed by quantitatively describing the deviation between the wind power prediction results and the actual power generation. The historical wind power prediction error sequences are used to characterize the degree of deviation between the wind power prediction results and the actual operating status in the time dimension. Statistical modeling of the error sequences can depict the distribution characteristics of wind power prediction errors under different time conditions. The historical wind power prediction error sequence is obtained by calculating the difference between the historical actual wind power data and the historical wind power prediction data of each wind power grid at the corresponding time point. The historical actual wind power data represents the real power output information of the wind turbine recorded by the power monitoring device during operation. The historical wind power prediction data represents the wind power prediction results calculated based on meteorological conditions, wind speed changes and prediction models. By calculating the difference between the two types of data at the same time point, the error value reflecting the degree of prediction deviation can be obtained, and they are arranged in chronological order to form the historical wind power prediction error sequence.
[0029] Furthermore, the prediction error feature vector extracted from the historical wind power prediction error sequence represents the statistical characteristics and variation patterns of the error data based on the error sequence. The prediction error feature vector is a multi-dimensional feature representation formed by combining multiple statistical indicators in the error sequence, used to characterize the distribution, fluctuation amplitude, and trend of the prediction error over time. The prediction error feature vector can include statistical features such as the error mean, error variance, skewness coefficient, kurtosis coefficient, and error rate of change. By extracting features from the error sequence, the original time series data can be transformed into a structured feature expression that can be used for probabilistic modeling, thereby improving the modeling stability and computational efficiency in the subsequent probability density estimation process.
[0030] Furthermore, based on the prediction error feature vector, the probability density of the prediction error sample set is estimated, and a prediction error probability density function is constructed. Statistical analysis methods are then used to continuously model the distribution of prediction error samples across different value intervals. The prediction error sample set represents the set of all prediction error data samples collected during the historical operating cycle. The probability density estimation method calculates the probability density distribution of error values across different intervals. The probability density function describes the probability distribution of prediction error variables in a continuous numerical space, revealing the probability trends of error values within various intervals. Subsequently, a prediction error probability distribution model is constructed for each wind power grid based on the prediction error probability density function. This model provides a probabilistic description of the uncertainty in wind power prediction errors and forms a statistical model structure reflecting the distribution characteristics of wind power grid prediction errors, thus providing fundamental probabilistic model support for wind power grid connection risk assessment and grid connection control decisions.
[0031] Based on the multiple prediction error probability distribution models, calculate multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids.
[0032] Furthermore, this application also includes: calculating multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids based on the multiple prediction error probability distribution models, wherein each set of grid connection risk probability vectors includes the probability of power exceeding the limit, the probability of power surge, and the probability of power drop when any wind power grid is connected to the grid.
[0033] Specifically, multiple sets of grid connection risk probability vectors corresponding to multiple wind power grids are calculated based on multiple prediction error probability distribution models. This represents the probabilistic quantification of the potential impact of prediction errors on grid operation after completing the statistical modeling of wind power prediction errors and analyzing the characteristics of the error probability distribution. The prediction error probability distribution model describes the probabilistic characteristics of wind power prediction errors occurring within different value ranges. By integrating the probability distribution function or performing probability interval analysis, the probability magnitude of prediction errors occurring in different intervals can be obtained. The grid connection risk probability vector represents a multidimensional set of probability parameters composed of multiple risk probability indicators, used to characterize the power fluctuation risk level that may be caused by wind power grids during grid connection operation. By calculating the prediction error probability distribution model corresponding to each wind power grid, a set of probability vectors reflecting the degree of grid connection risk under different operating conditions can be obtained, thereby achieving a quantitative description of the grid connection risk status of multiple wind power grids.
[0034] Furthermore, each grid connection risk probability vector includes power over-limit probability, power surge probability, and power drop probability, used to classify and characterize various risk scenarios during the wind power grid connection process. The power over-limit probability represents the probability level that the actual output power of the wind power grid exceeds the grid's allowable power range due to prediction errors during grid connection operation. Power over-limit conditions may lead to operational risks such as grid node overload or insufficient line transmission capacity. Therefore, probability calculations can reflect the likelihood of power exceeding limits during grid connection. The power surge probability represents the probability level of a rapid increase in the actual output power of wind power within a short period. Power surge conditions may cause momentary imbalances in grid power balance or voltage fluctuations. The power surge probability index can be obtained by statistically calculating the probability of the rising interval in the prediction error distribution. The power drop probability represents the probability level of a rapid decrease in the output power of wind power within a short period. Power drop conditions may increase the demand for grid reserve capacity or trigger frequency regulation pressure. The power drop probability index can be obtained by statistically analyzing the probability of the corresponding falling interval in the prediction error distribution. By combining the power over-limit probability, power surge probability, and power drop probability, a complete grid connection risk probability vector can be formed, thereby reflecting the grid connection uncertainty risk of wind power grid under different operating conditions from multiple risk dimensions.
[0035] Furthermore, the grid connection risk probability vector is used to characterize the volatility and uncertainty risks faced by wind power grids under different time points or different power output levels. Different time points correspond to the changing trends of wind power output and the changing states of meteorological conditions, while different power output levels correspond to the operating states of wind turbines in different power output ranges. By performing conditional probability analysis on the prediction error probability distribution model, risk probability results under different time scales or different power ranges can be obtained, thus forming multiple sets of grid connection risk probability vectors. Multiple sets of grid connection risk probability vectors can comprehensively reflect the power fluctuation risks that may arise from grid connection of wind power grids under different operating scenarios, providing a data foundation for the grid dispatching system to subsequently plan grid connection timing and manage risks.
[0036] Based on the multiple sets of grid connection risk probability vectors, a grid connection access time sequence window is generated for each wind power grid. Time sequence window overlap analysis is performed on the multiple grid connection access time sequence windows corresponding to the multiple wind power grids to obtain a set of wind power grid access pairs that meet the grid connection access conditions.
[0037] Furthermore, this application also includes: obtaining the grid-connected node capacity and grid-connected line margin; constructing a grid connection impact index assessment model based on the grid-connected node capacity and grid-connected line margin; inputting multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids into the grid connection impact index assessment model to calculate the grid connection impact index of each wind power grid; adaptively adjusting the grid connection access window width based on the grid connection impact index of each wind power grid to obtain an adaptive grid connection access window width; and generating a wind power grid grid connection access timing window based on the adaptive grid connection access window width of each wind power grid.
[0038] Furthermore, this application also includes: analyzing the stable moment of each wind power grid, wherein the stable moment is the moment when the predicted power change rate is less than a preset change rate threshold; and generating a wind power grid grid connection timing window with the stable moment of each wind power grid as the window center and the adaptive grid connection window width of each wind power grid.
[0039] Furthermore, this application also includes: aligning the multiple grid connection timing windows corresponding to the multiple wind power grids to obtain a wind power grid grid connection window set; constructing a wind power grid timing window topology map based on the wind power grid grid connection window set; performing grid connection conflict analysis on the wind power grid timing window topology map to obtain a wind power grid grid connection conflict matrix; and identifying a set of wind power grid access pairs that meet the grid connection conditions based on the wind power grid grid connection conflict matrix.
[0040] Furthermore, this application also includes: constructing a wind power grid timing window topology graph based on the wind power grid grid connection access window set, wherein the wind power grid timing window topology graph includes topology nodes and topology connection edges; the topology nodes are the wind power grid grid connection access windows in the wind power grid grid connection access window set, and the topology connection edges are established by determining whether the grid connection access timing windows of two wind power grids in the wind power grid grid connection access window set overlap.
[0041] Furthermore, this application also includes: traversing all candidate node pairs in the grid connection conflict matrix, when the grid connection windows of the two wind power grids of any candidate node pair overlap in time; determining whether the total capacity of the grid-connected nodes and the total margin of the grid-connected lines are both less than the corresponding preset thresholds; if both are less than the corresponding preset thresholds, marking the current candidate node pair as a wind power grid access pair, and outputting a set of wind power grid access pairs.
[0042] Specifically, the grid-connected node capacity and grid-connected line margin are obtained, and a grid-connected impact index assessment model is constructed based on these parameters. This model provides a parameterized description of the grid structure's operational capacity at the wind power grid connection location during grid-connected scheduling analysis. Grid-connected node capacity represents the maximum power capacity that a power node in the grid can carry under stable operating conditions for connecting to the wind power grid. This capacity is determined by a combination of the node's rated capacity, grid operational safety constraints, and power flow distribution conditions, reflecting the upper limit of the power scale that a node is allowed to carry under grid-connected operation. Grid-connected line margin represents the unused transmission capacity space of the transmission lines connecting the grid-connected node to the upstream grid or adjacent nodes under current operating conditions. The remaining available transmission capacity can be obtained by comparing the line's rated transmission capacity with the actual operating load level, reflecting the safety margin of the transmission lines when carrying additional grid-connected power. After obtaining the grid-connected node capacity and grid-connected line margin parameters, a grid-connected impact index assessment model is established to comprehensively evaluate the potential power disturbance impact when wind power grids are connected to the grid. The grid-connected impact index assessment model is used to integrate the grid structure carrying capacity and wind power fluctuation risk characteristics, thereby forming a quantitative evaluation mechanism for the degree of grid connection impact.
[0043] Furthermore, multiple sets of grid connection risk probability vectors corresponding to various wind power grids are input into the grid connection impact index assessment model to calculate the grid connection impact index for each wind power grid. This signifies that the operational uncertainties of wind power grids are incorporated into the comprehensive assessment system. The grid connection risk probability vector describes the probability of risks such as power over-limit, power surge, and power drop that may occur during the grid connection operation of wind power grids. By using different risk probability indicators as model input variables and jointly calculating them with grid connection node capacity and grid connection line margin parameters, the numerical result of the grid connection impact index can be obtained. The grid connection impact index is used to quantify the comprehensive impact that wind power grids may have on the stability of grid operation during grid connection. The index value reflects the matching relationship between wind power fluctuation risk and grid carrying capacity. By calculating the grid connection impact index for each wind power grid separately, a risk assessment index for grid connection scheduling decisions can be formed.
[0044] Furthermore, based on the grid connection impact index of each wind power grid, the width of the grid connection window is adaptively adjusted and an adaptive grid connection window width is obtained. This represents the control of grid connection risk by dynamically adjusting the grid connection time range after completing the quantitative assessment of grid connection risk. The grid connection window width represents the length of the time range during which the wind power grid is allowed to perform grid connection operations, used to describe the executable interval of the grid connection operation in the time dimension. Where W... i (t)=W max × CI i (t)= W i (t) represents the grid connection window width at time t, i.e., the window width at time t, W max Where e is the maximum allowed window, and CI is the base of the natural logarithm. i (t) represents the grid connection impact index, i.e., the impact index at time t, where t is a time variable. To provide grid connection adjustment margin, C i M represents the capacity of the grid-connected nodes. i Here, represents the grid connection margin, and 'i' denotes the index of the i-th wind power grid. A large grid connection impact index indicates a higher risk of disturbance to the grid from wind power grid connection. Reducing the grid connection window width can limit the time interval for grid connection operations, thereby mitigating the impact risk on grid operation. Conversely, a small grid connection impact index indicates a relatively smaller impact on grid operation from wind power grid connection. Expanding the grid connection window width can improve the flexibility of grid connection scheduling. The adaptive grid connection window width represents the time window range formed after dynamic adjustment based on the grid connection impact index, enabling differentiated configuration of grid connection scheduling strategies according to the risk levels of different wind power grids.
[0045] Furthermore, analyzing the stable moments of each wind power grid represents the time-specific identification of the output power variation state of the wind power grid. A wind power grid represents a collection of wind power generation units formed by several wind turbine generators through a local power grid structure. During operation, the wind power grid is affected by factors such as wind speed changes, meteorological conditions, and generator control strategies, resulting in certain fluctuations in output power. A stable moment represents the point in time when the power variation of the wind power grid is in a relatively stable state. By calculating the rate of change of the predicted power time series, the power variation amplitude between different time points can be obtained. The predicted power change rate represents the relative proportion or difference of the change in the predicted wind power value within adjacent time intervals, reflecting the speed of change in wind power output over a short period. When the predicted power change rate is less than a preset change rate threshold, it indicates that the wind power output variation amplitude is within a small range, and the wind power curve shows a relatively stable trend, thus the corresponding time point can be determined as a stable moment. The preset change rate threshold represents the boundary value of the change rate used to distinguish between power fluctuation states and power stability states. It is set through statistical analysis of wind power prediction data or based on grid operation experience to determine the identification conditions for stable operation states.
[0046] Furthermore, using the stable moment of each wind power grid as the window center and the adaptive grid connection window width of each wind power grid as the window width, a wind power grid grid connection timing window is generated. This indicates that after the stable moment is identified, the specific grid connection time interval is determined through the time window construction method. The window center represents a reference time point on the time axis as the middle position of the grid connection time interval. Using the stable moment as the window center ensures that the grid connection operation occurs during the operation phase when wind power changes are relatively smooth, thereby reducing the grid impact risk caused by power fluctuations during the grid connection process. The adaptive grid connection window width represents the length of the time range dynamically determined according to the grid connection risk level of the wind power grid. It is adjusted based on the grid connection impact index assessment results, so that different wind power grids have a time window range that matches their risk level when connecting to the grid. Under the condition that the stable moment is the time center point, a complete grid connection time interval can be formed by symmetrically expanding the adaptive grid connection window width on both sides of the time axis. The wind power grid grid connection timing window represents the time range structure used to limit the wind power grid to perform grid connection operations. The timing window expression can provide clear time constraints for subsequent multi-wind power grid grid connection coordination and scheduling. Among them, TW i =[t i -W i / 2,t i +W i / 2], TW i Let t represent the i-th time window. i For the steady moment, i.e. the center moment of the time window, W i The t represents the width of the grid connection window, i represents the index of the i-th wind power grid, and t represents the width of the grid connection window. i -W i / 2 represents the start time of the time window, t i +W i / 2 indicates the end time of the time window.
[0047] The process of aligning multiple grid-connection time windows corresponding to multiple wind power grids to obtain a set of wind power grid connection windows represents the process of applying a unified time reference to the grid-connection time information of different wind power grids after determining the grid-connection time intervals for each wind power grid. A grid-connection time window represents the time interval range on the time axis within which grid-connection operations are permitted. This time interval is typically determined by a start time and an end time, describing the executable range of the grid-connection operation in the time dimension. Time alignment means mapping multiple wind power grid connection time windows to a unified time coordinate system. By using a unified time scale, the grid-connection windows of different wind power grids are synchronized, enabling direct comparison and overlap analysis between the grid-connection time intervals of different wind power grids. After completing the time alignment process, a set of wind power grid connection windows containing information on multiple wind power grid connection time windows can be obtained. This set is used to centrally represent the grid-connection time interval information of all wind power grids within the power grid area.
[0048] Furthermore, a wind power grid time-series window topology diagram is constructed based on the wind power grid grid connection access window set, representing a structured modeling of the temporal relationships between different wind power grids using a graph structure approach. The wind power grid grid connection access window set represents a collection of multiple wind power grid grid connection access time-series window data aggregated under a unified time coordinate system. Each grid connection access time-series window describes the range of grid-connectable operation for the corresponding wind power grid on the time axis. The wind power grid time-series window topology diagram represents a network model structure that uses graph theory to express the relationships between multiple grid connection time intervals. The relationships between different wind power grid grid connection time windows can be intuitively described through nodes and connections. In the wind power grid time-series window topology diagram, topology nodes and topology connecting edges constitute the basic structural units. Topology nodes represent the entities of each grid connection time interval in the grid connection access window set, and topology connecting edges represent the temporal relationships between different grid connection time intervals. This graph structure representation method forms a complete grid connection time relationship network, thus providing a structured data foundation for grid connection coordination analysis.
[0049] Furthermore, the topology node represents the structural representation of the wind power grid connection windows in the set of wind power grid connection windows, used to clarify the specific objects corresponding to the node elements in the graph structure. A wind power grid connection window represents the time interval within which a wind power grid is allowed to perform grid connection operations on the time axis, defined by start and end time parameters. In the power grid dispatch system, it is used to limit the time boundaries of wind power grid connection behavior. By mapping each connection window to a node in the topology graph, a unified expression of multiple wind power grid connection time intervals can be achieved in the graph structure. Meanwhile, the topology connection edges are established by determining whether there is a time overlap between the grid connection time windows of two wind power grids in the set of wind power grid connection access windows. When the two grid connection time windows have overlapping intervals on the time axis, a connection relationship is established between the corresponding topology nodes to indicate the possibility or potential coordination relationship between the two wind power grids in terms of grid connection time. When the two grid connection time windows do not have overlapping intervals on the time axis, no topology connection relationship is established, so that the topology structure can accurately reflect the overlapping characteristics of the grid connection time windows between different wind power grids.
[0050] Furthermore, grid connection conflict analysis is performed on the wind power grid time-series window topology diagram, and a wind power grid connection conflict matrix is obtained. This represents the identification of potential grid connection conflict relationships between different wind power grids based on the topology structure. Grid connection conflict indicates a state where multiple wind power grids have overlapping grid connection demands in a time interval and may simultaneously have a superimposed impact on grid operation. By analyzing the connection relationships between nodes in the topology diagram, combinations of grid connection windows with temporal overlap can be identified. After completing the conflict relationship identification, the conflict relationship is formally expressed through a matrix structure. The wind power grid connection conflict matrix represents a data structure that records the grid connection relationships of different wind power grids in matrix form. The rows and columns in the matrix correspond to different wind power grids or corresponding grid connection windows, respectively. The matrix element values are used to indicate whether there is a grid connection conflict relationship between two corresponding wind power grids. The matrix structure enables a systematic description of the grid connection conflict state of multiple wind power grids.
[0051] Traversing all candidate node pairs in the grid connection conflict matrix and determining whether the grid connection windows of the two wind power grids for any candidate node pair overlap in time represents the step-by-step analysis of node combinations recorded in the conflict matrix after completing the grid connection conflict relationship modeling. The grid connection conflict matrix is a matrix-based data structure used to record the grid connection time relationships and potential conflict states between different wind power grids. The rows and columns of the matrix correspond to the grid connection windows of different wind power grids within the power grid area, and the matrix elements are used to identify the grid connection time relationship or conflict state between corresponding two wind power grids. A candidate node pair represents a combination unit consisting of two wind power grid nodes in the grid connection conflict matrix, used as the basic judgment object for grid connection coordination analysis. The traversal operation represents accessing and analyzing all candidate node combinations one by one according to the matrix structure, thereby ensuring that all possible wind power grid combinations can be included in the grid connection coordination judgment scope. During the traversal, by comparing the time intervals of the grid connection windows corresponding to the two wind power grids, it can be determined whether there is an overlapping interval between the two windows on the time axis. When the two grid connection windows overlap on the time axis, it indicates that the corresponding wind power grids may perform grid connection operations simultaneously within the same time period, thus requiring further constraint judgment on the grid carrying capacity.
[0052] Furthermore, after confirming that the candidate nodes have overlapping time windows for connecting to the corresponding two wind power grids, a threshold comparison is performed on the total capacity of the grid-connected nodes and the total margin of the grid-connected lines to assess whether the grid's carrying capacity within the stated time interval meets the grid-connected operation conditions. The total capacity of the grid-connected nodes represents the overall power access capacity that the grid-connected nodes used to connect to the wind power grid can carry under the current operating conditions. This capacity index is formed by comprehensively considering the rated capacity of the node equipment, grid operation safety constraints, and system power flow distribution. The total margin of the grid-connected lines represents the remaining available transmission capacity of the transmission lines connecting the grid-connected nodes to other nodes in the grid under the current operating conditions. The remaining transmission capacity can be obtained by comparing the rated transmission capacity of the lines with the real-time operating load level. The preset threshold represents a capacity limit parameter pre-set according to grid operation safety specifications and dispatch management requirements, used as a basis for judging the feasibility of grid-connected operation. When the total capacity of grid-connected nodes and the total margin of grid-connected lines are both less than the corresponding preset thresholds, it indicates that the power grid still has sufficient capacity to support the simultaneous operation of two wind power grids within the corresponding time interval. In this case, the current candidate node pair is marked as a wind power grid access pair, representing that the two wind power grids can perform grid connection operations within the same time window. By traversing and judging all candidate node pairs, a set of wind power grid access pairs consisting of multiple wind power grid access pairs can be output, representing the wind power grid combination relationship that can simultaneously perform grid connection operations under grid connection time coordination conditions.
[0053] Based on the wind power grid access set, a grid connection control command is generated, and the grid connection access sequence of each wind power grid is controlled by the grid connection control command.
[0054] Specifically, the process of generating grid-connected control commands based on the set of wind power grid access pairs represents the formation of executable scheduling and control information based on the combination relationships of wind power grids that meet the grid connection conditions after completing the wind power grid grid connection coordination analysis. The set of wind power grid access pairs represents a data set composed of multiple wind power grid access pairs. Each wind power grid access pair represents the combination relationship where two wind power grids can simultaneously perform grid connection operations under time interval and grid carrying capacity constraints. The set of wind power grid access pairs is formed by summarizing all combination relationships that meet the grid connection constraints. The grid-connected control command represents the operation instruction information generated by the grid dispatch and control system and issued to the wind power grid control equipment or grid connection control device. It specifies the specific time arrangement and control strategy for the wind power grid to perform grid connection operations. The grid-connected control command typically includes grid connection time parameters, grid connection sequence information, and relevant operational constraint parameters. By transforming the grid connection coordination results into executable control information in the form of commands, unified scheduling and management of wind power grid grid connection operations can be achieved.
[0055] Furthermore, the process of controlling the grid connection sequence of various wind power grids using grid connection control commands represents the power grid dispatching system's implementation of time coordination and operational control over the wind power grid connection process based on generated control commands. Grid connection sequence refers to the order in which different wind power grids perform grid connection operations and the corresponding time intervals. By rationally arranging the grid connection sequence, excessive power fluctuations from multiple wind power grids occurring simultaneously can be avoided. The control process represents the unified coordination and management of wind power grid grid connection behavior using the dispatching control system. Control commands constrain wind power grids to perform grid connection operations within specified time windows, while simultaneously controlling the grid connection sequence and time, ensuring that the grid connection behavior of multiple wind power grids is implemented in an orderly manner according to a pre-planned time structure. Grid connection sequence control can reduce the impact of wind power output fluctuations on power grid operational stability and improve the power grid's coordination capability for the connection of multiple wind power grids.
[0056] Furthermore, this application also includes: obtaining multiple historical wind power prediction error sequences for each wind power grid within the power grid area at multiple time scales; performing multi-scale optimization of the prediction error probability distribution model corresponding to each wind power grid based on the multiple historical wind power prediction error sequences at multiple time scales; wherein, by selecting a time scale, multiple sets of grid connection risk probability vectors output by the prediction error probability distribution model are obtained at the corresponding time scale, and a grid connection access timing window for each wind power grid at the corresponding time scale is generated based on the multiple sets of grid connection risk probability vectors.
[0057] Specifically, the process of acquiring multiple historical wind power prediction error sequences for each wind power grid within a power grid area at multiple time scales represents the systematic collection and processing of error data under different time resolution conditions during the wind power prediction error modeling stage. The power grid area refers to the power supply and transmission network range uniformly managed by the power grid dispatch system, within which multiple wind power grids are distributed to transmit wind power to the grid. A wind power grid refers to a wind power generation unit composed of several wind turbine generators and local collection networks, exchanging power with the main power grid through grid-connected nodes. The time scale refers to the time unit used to describe the granularity of time data sampling or statistical analysis, such as minute-level, ten-minute-level, and hour-level time scales. Different time scales can reflect different fluctuation characteristics of wind power changes. The historical wind power prediction error sequence represents time series data formed by arranging the error values obtained by comparing actual wind power data and predicted wind power data within a historical operating cycle in chronological order. By acquiring prediction error sequences under multiple time scale conditions, the changing patterns of wind power prediction errors can be described from different time resolution perspectives.
[0058] Furthermore, the process of multi-scale optimization of the prediction error probability distribution model for each wind power grid based on multiple historical wind power prediction error sequences across multiple time scales represents the improvement and correction of the error probability model through statistical modeling methods after obtaining error data at different time resolutions. The prediction error probability distribution model represents a statistical model structure used to describe the probability distribution characteristics of wind power prediction errors within different value ranges; the probability distribution function reflects the overall pattern of error fluctuations. Multi-scale optimization involves comprehensively adjusting and optimizing the parameters of the probability distribution model by combining error sequence data from multiple time scales, enabling the model to adapt to error distribution characteristics under different time resolution conditions simultaneously. Through multi-scale optimization, the applicability of the prediction error probability distribution model under different time-varying conditions can be enhanced, improving the model's accuracy in describing the uncertainty of wind power fluctuations.
[0059] Furthermore, by selecting a time scale, multiple sets of grid connection risk probability vectors output by the prediction error probability distribution model are obtained at the corresponding time scale. The process of generating a grid connection access timing window for each wind power grid at the corresponding time scale based on these multiple sets of grid connection risk probability vectors represents grid connection risk assessment and grid connection time planning based on multi-scale modeling and different time resolutions. Time scale selection means choosing a specific time resolution for analysis and calculation based on grid dispatch requirements or wind power fluctuation characteristics. For example, focusing on the risk of rapid power fluctuations at short time scales and on power change trends at longer time scales. The prediction error probability distribution model can output corresponding probability distribution results at the selected time scale. By calculating risk indicators from the probability distribution, multiple sets of grid connection risk probability vectors can be obtained. These vectors represent risk indicators such as the probability of power exceeding limits, power surge, and power drop that may occur during the grid connection operation of the wind power grid. Through comprehensive analysis of the risk probability vectors, a grid connection access timing window suitable for the time resolution conditions can be generated at the corresponding time scale, describing the time interval range on the time axis where the wind power grid is allowed to perform grid connection operations.
[0060] In summary, the wind power prediction-assisted wind power grid connection management method provided in this application has the following technical effects: by achieving the technical goal of probabilistic modeling of the uncertainty of wind power prediction error and quantitative assessment of the grid connection risk of wind power grid, it achieves the technical effect of improving the scientific nature of wind power grid connection timing planning and enhancing the grid's operational stability and scheduling coordination capabilities in multi-wind power grid connection scenarios.
[0061] Example 2: Based on the same inventive concept as the wind power forecast-assisted wind power grid connection management method in the foregoing examples, this application also provides a wind power forecast-assisted wind power grid connection management device. Please refer to the appendix. Figure 2 The system includes: a prediction error probability distribution model construction module 1, used to acquire historical wind power prediction error sequences of multiple wind power grids within the power grid area, and construct multiple prediction error probability distribution models corresponding to the multiple wind power grids based on the historical wind power prediction error sequences; a grid connection risk probability vector calculation module 2, used to calculate multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids based on the multiple prediction error probability distribution models; a wind power grid access pair set acquisition module 3, used to generate a grid connection access time sequence window for each wind power grid based on the multiple sets of grid connection risk probability vectors, perform time sequence window overlap analysis on the multiple grid connection access time sequence windows corresponding to the multiple wind power grids, and obtain a set of wind power grid access pairs that meet the grid connection access conditions; and a control module 4, used to generate grid connection control instructions based on the set of wind power grid access pairs, and control the grid connection access time sequence of each wind power grid with the grid connection control instructions.
[0062] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used for: wherein the historical wind power prediction error sequence is obtained by calculating the error between the historical actual wind power data and the historical wind power prediction power data of each wind power grid at the corresponding time point; extracting the prediction error feature vector of the historical wind power prediction error sequence; performing probability density estimation on the prediction error sample set based on the prediction error feature vector, constructing a prediction error probability density function, and constructing a prediction error probability distribution model for each wind power grid based on the prediction error probability density function.
[0063] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used to: calculate multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids according to the multiple prediction error probability distribution models, wherein each set of grid connection risk probability vectors includes the probability of power exceeding the limit, the probability of power surge, and the probability of power drop when any wind power grid is connected to the grid.
[0064] Furthermore, the wind power prediction-assisted grid connection management device is also used for: obtaining the grid connection node capacity and grid connection line margin; constructing a grid connection impact index assessment model based on the grid connection node capacity and grid connection line margin; inputting multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids into the grid connection impact index assessment model to calculate the grid connection impact index of each wind power grid; adaptively adjusting the grid connection access window width based on the grid connection impact index of each wind power grid to obtain the adaptive grid connection access window width; and generating a wind power grid connection access timing window based on the adaptive grid connection access window width of each wind power grid.
[0065] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used to: analyze the stable moment of each wind power grid, wherein the stable moment is the moment when the predicted power change rate is less than a preset change rate threshold; and generate a wind power grid connection timing window with the stable moment of each wind power grid as the window center and the adaptive grid connection window width of each wind power grid.
[0066] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used for: time-aligning the multiple grid connection access time-series windows corresponding to the multiple wind power grids to obtain a wind power grid grid connection access window set; constructing a wind power grid time-series window topology map based on the wind power grid grid connection access window set; performing grid connection access conflict analysis on the wind power grid time-series window topology map to obtain a wind power grid grid connection access conflict matrix; and identifying a set of wind power grid access pairs that meet the grid connection access conditions according to the wind power grid grid connection access conflict matrix.
[0067] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used to: construct a wind power grid time-series window topology diagram based on the wind power grid grid connection access window set, wherein the wind power grid time-series window topology diagram includes topology nodes and topology connection edges; the topology nodes are the wind power grid grid connection access windows in the wind power grid grid connection access window set, and the topology connection edges are established by determining whether the grid connection access time-series windows of two wind power grids in the wind power grid grid connection access window set overlap.
[0068] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used to: traverse all candidate node pairs in the grid connection conflict matrix; when the grid connection windows of the two wind power grids of any candidate node pair overlap in time; determine whether the total capacity of the grid-connected nodes and the total margin of the grid-connected lines are both less than the corresponding preset thresholds; if both are less than the corresponding preset thresholds, mark the current candidate node pair as a wind power grid access pair and output the set of wind power grid access pairs.
[0069] Furthermore, the wind power prediction-assisted wind power grid connection management device is also used to: acquire multiple historical wind power prediction error sequences for each wind power grid within the power grid area at multiple time scales; optimize the prediction error probability distribution model corresponding to each wind power grid at multiple scales based on the multiple historical wind power prediction error sequences at multiple time scales; wherein, by selecting a time scale, multiple sets of grid connection risk probability vectors output by the prediction error probability distribution model are acquired at the corresponding time scale, and a grid connection access timing window for each wind power grid at the corresponding time scale is generated based on the multiple sets of grid connection risk probability vectors.
[0070] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The wind power prediction-assisted wind power grid connection management method and specific examples in the aforementioned embodiment one are also applicable to the wind power prediction-assisted wind power grid connection management device in this embodiment. Through the foregoing detailed description of the wind power prediction-assisted wind power grid connection management method, those skilled in the art can clearly understand the wind power prediction-assisted wind power grid connection management device in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.
[0071] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
[0072] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.
Claims
1. A wind power grid connection management method assisted by wind power forecasting, characterized in that, The method includes: Obtain historical wind power prediction error sequences of multiple wind power grids within the power grid area, and construct multiple prediction error probability distribution models corresponding to the multiple wind power grids based on the historical wind power prediction error sequences. Calculate multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids based on the multiple prediction error probability distribution models; Based on the multiple sets of grid connection risk probability vectors, a grid connection access time sequence window is generated for each wind power grid. Time sequence window overlap analysis is performed on the multiple grid connection access time sequence windows corresponding to the multiple wind power grids to obtain a set of wind power grid access pairs that meet the grid connection access conditions. Based on the wind power grid access set, a grid connection control command is generated, and the grid connection access sequence of each wind power grid is controlled by the grid connection control command; The method for generating the grid connection timing window for each wind power grid based on the multiple sets of grid connection risk probability vectors includes: Obtain the grid-connected node capacity and grid-connected line margin, and construct a grid-connected impact index assessment model based on the grid-connected node capacity and grid-connected line margin; The grid connection risk probability vectors corresponding to the multiple wind power grids are input into the grid connection impact index assessment model to calculate the grid connection impact index of each wind power grid; The grid connection window width is adaptively adjusted based on the grid connection impact index of each wind power grid to obtain the adaptive grid connection window width. The wind power grid connection timing window is generated based on the adaptive grid connection window width of each wind power grid. Perform time window overlap analysis on multiple grid connection time windows corresponding to the multiple wind power grids to obtain a set of wind power grid access pairs that meet the grid connection conditions. The method includes: The multiple grid connection timing windows corresponding to the multiple wind power grids are time-aligned to obtain a set of wind power grid connection windows. Construct a wind power grid time-series window topology diagram based on the aforementioned wind power grid grid connection access window set; Grid connection conflict analysis is performed on the time-series window topology diagram of the wind power grid to obtain the grid connection conflict matrix of the wind power grid; The set of wind power grid access pairs that meet the grid connection conditions is identified based on the wind power grid connection conflict matrix.
2. The wind power prediction-assisted grid connection management method as described in claim 1, characterized in that, Based on the historical wind power prediction error sequence, multiple prediction error probability distribution models are constructed for the multiple wind power grids. include: The historical wind power prediction error sequence is obtained by calculating the error between the historical actual wind power data and the historical wind power prediction data of each wind grid at the corresponding time point. Extract the prediction error feature vector from the historical wind power prediction error sequence; Based on the prediction error feature vector, the probability density of the prediction error sample set is estimated, a prediction error probability density function is constructed, and a prediction error probability distribution model for each wind power grid is constructed based on the prediction error probability density function.
3. The wind power prediction-assisted grid connection management method as described in claim 1, characterized in that, Based on the multiple prediction error probability distribution models, multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids are calculated. Each set of grid connection risk probability vectors includes the probability of power exceeding the limit, the probability of power surge, and the probability of power drop when any wind power grid is connected to the grid.
4. The wind power prediction-assisted grid connection management method as described in claim 1, characterized in that, The method for generating a wind power grid connection timing window based on the adaptive grid connection window width for each wind power grid includes: Analyze the stable moment of each wind power grid, where the stable moment is the moment when the predicted power change rate is less than a preset change rate threshold; Using the stable moment of each wind power grid as the center of the window and the adaptive grid connection window width of each wind power grid, a wind power grid grid connection timing window is generated.
5. The wind power prediction-assisted grid connection management method as described in claim 1, characterized in that, A wind power grid time-series window topology graph is constructed based on the set of wind power grid connection windows, wherein the wind power grid time-series window topology graph includes topology nodes and topology connection edges; The topology node is the wind power grid connection window in the wind power grid connection window set, and the topology connection edge is established by determining whether the connection timing windows of two wind power grids in the wind power grid connection window set overlap. 6.The wind power prediction aided wind power grid connection management method according to claim 1, wherein, The method for identifying a set of wind power grid access pairs that meet the grid connection conditions based on the wind power grid connection conflict matrix includes: Traverse all candidate node pairs in the grid connection conflict matrix, when the grid connection windows of the two wind power grids of any candidate node pair overlap in time; Determine whether the total capacity of the grid-connected nodes and the total margin of the grid-connected lines are both less than the corresponding preset thresholds. If both are less than the corresponding preset thresholds, mark the current candidate node pair as a wind power grid access pair and output the set of wind power grid access pairs.
7. The wind power prediction-assisted grid connection management method as described in claim 1, characterized in that, The method includes: Multiple historical wind power prediction error sequences for each wind power grid within the power grid area at multiple time scales were obtained. Based on the multiple historical wind power prediction error sequences at the multiple time scales, the prediction error probability distribution model corresponding to each wind power grid is optimized at multiple scales. Specifically, by selecting a time scale, multiple sets of grid connection risk probability vectors output by the prediction error probability distribution model are obtained at the corresponding time scale, and grid connection access timing windows for each wind power grid at the corresponding time scale are generated based on the multiple sets of grid connection risk probability vectors.
8. A wind power grid connection control device assisted by wind power forecasting, characterized in that, The steps for implementing the wind power prediction-assisted wind power grid connection management method according to any one of claims 1 to 7 include: The prediction error probability distribution model construction module is used to obtain the historical wind power prediction error sequence of multiple wind power grids in the power grid area, and construct multiple prediction error probability distribution models corresponding to the multiple wind power grids based on the historical wind power prediction error sequence. The grid connection risk probability vector calculation module is used to calculate multiple sets of grid connection risk probability vectors corresponding to the multiple wind power grids based on the multiple prediction error probability distribution models. The wind power grid access pair set acquisition module is used to generate a grid connection time sequence window for each wind power grid based on the multiple sets of grid connection risk probability vectors, perform time sequence window overlap analysis on the multiple grid connection time sequence windows corresponding to the multiple wind power grids, and obtain a set of wind power grid access pairs that meet the grid connection conditions. The control module is used to generate grid connection control instructions based on the set of wind power grid access pairs, and to control the grid connection timing of each wind power grid using the grid connection control instructions.