Multi-time scale wind-light-charge extreme scene extraction method and system and storage medium

Abstract

The invention discloses a multi-time scale wind-light-charge extreme scene extraction method and system and a storage medium. The method comprises the following steps: acquiring hourly time sequence data of each node in a power grid in horizontal annual wind-light-load as scene data to form a time sequence scene; clustering the time sequence scene in a time dimension to obtain a plurality of typical wind-light-load operation scenes based on combination of time correlation and geographical distribution characteristics; and on the basis of each typical wind-light-load operation scene, taking a time sequence scene which corresponds to the typical wind-light-load operation scene and is farthest from the clustering center as an extreme scene corresponding to the typical wind-light-load operation scene. According to the method, the time correlation and geographical distribution characteristics among wind-light-load are considered through a weighted clustering method, so that the extreme operation scene of the system is identified, the stability and safety of system planning are evaluated, and personnel are assisted in evaluating the operation safety of a system planning scheme and performing power grid stability scene analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of power system scene extraction, and in particular to a multi-time scale wind-solar-load extreme scene extraction method, system and storage medium. Background Art

[0002] Against the backdrop of deepening consensus on addressing global climate change and promoting carbon emission reduction, the utilization rate of renewable energy such as wind and solar energy has increased significantly. Although these forms of energy have little impact on the environment, their power generation is highly volatile, which increases the complexity of power system operation. In addition, the volatility of electricity demand, affected by multiple factors such as seasonality, climatic conditions, and socio-economic activities, together with the uncertainty of renewable energy supply, challenges the stable operation of the power grid. As climate change leads to an increase in the frequency and intensity of extreme weather events, these events may have a significant impact on the balance of electricity supply and demand. Therefore, in order to maintain the safety and reliability of the power system, it is critical to identify and prepare for extreme scenarios in wind and solar power generation and electricity demand to ensure the stability and response capabilities of the power system in the face of uncertain challenges.

[0003] Existing research on wind-solar-load extreme scenario extraction mainly focuses on using statistical methods, machine learning techniques and system simulation to predict and analyze extreme changes in renewable energy generation (especially wind and solar energy) and power load. These studies are committed to improving the adaptability and resilience of power systems to renewable energy fluctuations and uncertainties, and are an important part of the research field of power system reliability and stability. However, existing models have limited ability to deal with the uncertainty and variability of extreme events, and it is difficult to fully cover all potential extreme situations. The generalization ability of the model is also a problem. The specific conditions of climate and power systems in different regions will make it impossible to directly migrate and apply the model. Summary of the invention

[0004] In view of the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide a multi-time-scale wind-solar-load extreme scenario extraction method, and to study the methods for determining the wind-solar-load clustering feature vectors and the basis for judging extreme scenarios for extreme scenarios such as long-term low output of new energy, long-term high output, and load surge or drop. In addition, for the two scenarios of power supply guarantee and consumption guarantee, a wind-solar-load extreme scenario extraction method based on extreme value extrapolation is studied.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A multi-time scale wind-solar-load extreme scene extraction method comprises the following steps:

[0007] Obtain the hourly time series data of wind, light and load of each node in the power grid in the normal year as scenario data to form a time series scenario;

[0008] Clustering the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics;

[0009] Based on each typical wind-solar-load operation scenario, the time sequence scenario corresponding to the typical wind-solar-load operation scenario and farthest from the cluster center is used as the extreme scenario corresponding to the typical wind-solar-load operation scenario.

[0010] As a preferred solution, the time series scenarios are clustered in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics, specifically including:

[0011] According to the time series data of each wind-solar-load variable in the time series scenario, weight the wind-solar-load variable, wherein the wind-solar-load variable includes wind power, photovoltaic power and load of each node in the power grid;

[0012] According to the weights of the wind-solar-load variables, the time series scenarios are clustered based on weighted Euclidean distance to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics.

[0013] As a preferred solution, the weighting of the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario specifically includes:

[0014] According to the time series data of each wind-solar-load variable in the time series scenario, the wind-solar-load variable is weighted by using the entropy weight method to obtain the objective weight of each variable;

[0015] According to the capacity of each wind-solar-load variable, the subjective weight of each variable is calculated;

[0016] The objective weights and subjective weights of the variables are weighted averaged to obtain the combined weights of the variables, and the combined weights are used as the weights of the wind-solar-load variables.

[0017] As a preferred solution, the specific method of determining the wind-solar-load variable weights by using the entropy weight method according to the time series data of each wind-solar-load variable in the time series scenario and obtaining the objective weights of each variable is as follows:

[0018] Assume that there are T time series scene objects to be clustered in the time dimension, and each of the time series scenes contains N WD +N Load +N PV dimensional variables, and define each variable as a characteristic index, calculate the entropy value of each characteristic index, and determine the objective weight of each variable according to the entropy value of each characteristic index. The method for obtaining the objective weight of each variable is as follows:

[0019]

[0020] In the formula, H i is the entropy value of the i-th feature index, n is the total dimension of the time series variable of each scene, T is the total number of time series scene objects, and f ij is the proportion of the jth temporal scene object at the i-th feature index, s ij is the data of the variable, p i is the entropy weight of the ith feature index, N WD 、N Load 、N PV are the dimensional variables of each time series scenario.

[0021] As a preferred solution, the specific method of performing weighted averaging of the objective weight and subjective weight of each variable to obtain the combined weight of each variable is:

[0022]

[0023] In the formula, w i is the combined weight of each variable, p i is the objective weight of the i-th characteristic index determined by the entropy weight method, q i is the subjective weight of the determined i-th characteristic index.

[0024] As a preferred solution, the calculation formula of the weighted Euclidean distance is as follows:

[0025]

[0026] In the formula, x i is the i-th element in the time series scene matrix dataset, x j is the jth element in the time series scene dataset, d(x i ,x j ) is x i and x j The weighted Euclidean distance between v is the weight of the v-th dimension variable of the time series scene data, h is the dimension of the time series scene data, x iv is the value of the ith data on the vth variable, x jv is the value of the j-th data on the v-th variable.

[0027] As a preferred solution, based on each typical wind-solar-load operation scenario, the time sequence scenario farthest from the cluster center corresponding to the typical wind-solar-load operation scenario is used as the extreme scenario corresponding to the typical wind-solar-load operation scenario, specifically including:

[0028] For each typical wind-solar-load operation scenario, according to the preset required number, from the class corresponding to the typical wind-solar-load operation scenario, among the time series scenes with the farthest weighted Euclidean distance, a preset number of time series scenes farthest from the cluster center are selected as the extreme scenes corresponding to the typical wind-solar-load operation scenario.

[0029] Accordingly, the present invention also provides a multi-time scale wind-solar-load extreme scene extraction system, comprising: a data acquisition module, a typical scene module and an extreme scene module;

[0030] The data acquisition module is used to acquire the hourly time series data of wind, light and load of each node in the power grid in a normal year as scene data to form a time series scene;

[0031] The typical scenario module is used to cluster the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics;

[0032] The extreme scenario module is used to take the time sequence scenario farthest from the cluster center corresponding to each typical wind-solar-load operation scenario as the extreme scenario corresponding to the typical wind-solar-load operation scenario based on each typical wind-solar-load operation scenario.

[0033] In the above multi-time scale wind-solar-load extreme scene extraction system, as a preferred solution, the typical scene module includes: a weighting unit and a typical scene unit;

[0034] The weighting unit is used to weight the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario, and the wind-solar-load variables include wind power, photovoltaic power and load of each node in the power grid;

[0035] The typical scenario unit is used to cluster the time series scenarios based on the weighted Euclidean distance according to the weights of the wind-solar-load variables, so as to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics.

[0036] In the above multi-time scale wind-solar-load extreme scene extraction system, as a preferred solution, the weighting unit includes: an objective weighting subunit, a subjective weighting subunit and a combined weighting subunit;

[0037] The objective weighting subunit is used to weight the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario by using the entropy weight method to obtain the objective weight of each variable, wherein the wind-solar-load variable includes wind power, photovoltaic power and load of each node in the power grid;

[0038] The subjective weighting subunit is used to calculate the subjective weight of each variable according to the capacity of each wind-solar-load variable;

[0039] The combined weighting subunit is used to perform weighted averaging on the objective weighting subunit and the subjective weighting subunit to obtain the combined weight of each variable.

[0040] In a third aspect, the present invention further provides a storage medium comprising a computer executable program, which, when executed by a computer processor, is used to execute the multi-time-scale wind-light-load extreme scene extraction method described in any one of the preceding items.

[0041] Compared with the prior art, the present invention has the following technical effects:

[0042] The present invention proposes a weighted clustering method to consider the time correlation and geographical distribution characteristics between wind, light and load, so as to quickly identify the extreme operation scenarios of the system to evaluate the stability and safety of system planning. Specifically, the time series data of wind, light and load at each node in the power grid every hour in the horizontal year is used as the time series scenario, and a combined weighted clustering method is adopted to select the typical wind-light-load operation scenario of the power grid, and extract the edge operation points away from the typical wind-light-load operation scenario corresponding to the distance from the cluster center as the extreme scenario. The clustering process takes into account the fluctuations of different wind farms and loads and the impact of capacity on extreme scenarios, thereby further improving the efficiency and rationality of extreme scenario extraction. This method can find large-scale wind and light in the planning stage and conduct stability assessment and flow analysis on the weak links of power grid operation, so that planners can quickly evaluate the operating safety of system planning schemes and improve the rationality and scientificity of large-scale power grid stability scenario analysis.

[0043] The present invention further extracts the setting of different weight values ​​in clustering according to the volatility of different variables over time and their impact on extreme scenarios, determines the objective and subjective weight ratios of the variables by utilizing the entropy weight method and the capacity method, and determines the combined weights of the variables by weighted averaging them, performs scene clustering by weighted Euclidean distance, and takes the farthest scene point as the extreme scene, thereby improving the rationality of scene extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to make the purpose, technical solution and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0045] Figure 1 A flow chart of a multi-time scale wind-solar-load extreme scene extraction method provided by the present invention;

[0046] Figure 2 is a flow chart of the weighting process of the variables of the present invention;

[0047] Figure 3 A diagram of an embodiment of an extreme scene extraction method provided by the present invention;

[0048] Figure 4 The normalized data of wind-solar-load and the extreme scene diagram of the embodiment of the present invention;

[0049] Figure 5 The wind, solar and load distribution diagrams of the distributed weighted clustering and the centralized weighted clustering according to the embodiment of the present invention are shown;

[0050] Figure 6 Graph showing the relationship between the number of clusters and the extreme point square error according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but only represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0052] The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0053] Example:

[0054] The present invention provides a multi-time scale wind-light-load extreme scene extraction method. Figure 1 As shown, the specific steps include:

[0055] S1. Obtain the hourly time series data of wind, light and load of each node in the power grid in the horizontal year as scenario data to form a time series scenario;

[0056] S2. Clustering the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics;

[0057] S3. Based on each typical wind-solar-load operation scenario, the time series scenario corresponding to the typical wind-solar-load operation scenario and farthest from the cluster center is used as the extreme scenario corresponding to the typical wind-solar-load operation scenario.

[0058] In this embodiment, most of the existing wind-solar-load extreme scene extraction methods are carried out for one type of wind, solar, and load, and there are few studies on wind-solar-load at the same time. In addition, in the process of generating time series, the influencing factors are not considered comprehensively, for example: season, date type, meteorological factors, etc. are not considered enough. This embodiment fully considers the correlation between wind-solar-load, and takes into account a variety of influencing factors represented by meteorology, and generates multi-time scale wind-solar-load extreme scene extraction, which can greatly improve the accuracy of extreme scene extraction and serve the operation and planning of the power system.

[0059] The present invention provides a multi-time scale wind-solar-load extreme scene extraction method based on weighted clustering, which specifically comprises the following steps:

[0060] S1. Obtain the hourly wind-solar-load time series data of each node in the power grid in the horizontal year as scenario data, and construct a time series scenario set; that is:

[0061] S11. Construct a set of combined wind-solar-load operation scenarios based on time correlation and geographical distribution characteristics.

[0062] S11 is specifically to construct a combined wind-solar-load operation scenario set based on time correlation and geographical distribution characteristics, which can use the annual hourly time series data of node wind power, node photovoltaic and node load levels.

[0063] The variable matrix for studying the hourly time series scenario is constructed as follows:

[0064] P j =(P j1 , P j2 , ..., P jT );

[0065] Where matrix, scene P j is the wind and solar power output or load of the jth node, with a total of N nodes, where j = 1, 2, 3, ..., N, T is the total number of scenarios, and generally 1 hour is one scenario. In the research year, T can be set to 8760 scenarios, P jT is the wind and solar power output or load of the jth node in the Tth scenario.

[0066] S2. Cluster the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on the combination of time correlation and geographical distribution characteristics, such as Figure 2 As shown, including:

[0067] S21. According to the time series data of each wind-solar-load variable in the time series scenario, weight the wind-solar-load variable, wherein the wind-solar-load variable includes wind power, photovoltaic power and load of each node in the power grid; step S21 specifically includes:

[0068] S210, according to the time series data of each wind-solar-load variable in the time series scenario, weight the wind-solar-load variable using an entropy weight method to obtain an objective weight of each variable;

[0069] S211. Calculate the subjective weight of each variable according to the capacity of each wind-solar-load variable;

[0070] S212: Perform weighted averaging on the objective weights and subjective weights of the variables to obtain a combined weight of the variables, and use the combined weight as the weight of each wind-solar-load variable.

[0071] In this embodiment, the types, dimensional differences and importance of clustering variables are significant, and clustering weights need to be set to reflect the characteristics of each variable. The traditional Euclidean distance fails to fully reflect the importance of variables in each dimension. For example, the contribution of wind and solar power generation variables in different locations to extreme scenarios will vary depending on their impact on the research problem. Therefore, in the clustering process, the importance of variables is different, and subjective weights and objective weights need to be considered. In terms of objective weights, variables with large time fluctuations often have a greater impact on the extraction of extreme scenes. The greater the variable fluctuation, the farther the data edge point is from the cluster center point, and the higher the possibility of contributing to the extreme edge point. In terms of subjective weights, it is necessary to analyze the impact of each variable on the research problem and give larger weights to variables with greater influence.

[0072] This embodiment uses the entropy weight method to assign weights to the wind-solar-load variables to obtain the objective weights of each variable. The basic principle is to measure the differences between variables by entropy values ​​and set the weights of each variable. It only reflects the volatility of variable data and has nothing to do with the type and relationship of the variables. According to the characteristics of the entropy weight method, entropy can be used to determine the degree of discreteness of variables. Based on this principle, calculating the Euclidean distance between scenes with entropy weight variables can obtain more accurate results.

[0073] In step S210, according to the time series data of each wind-solar-load variable in the time series scenario, the entropy weight method is used to determine the wind-solar-load variable weight, and the specific method of obtaining the objective weight of each variable is:

[0074] Assume that there are T time series scene objects to be clustered in the time dimension, and each of the time series scenes contains N WD +N Load +N PV dimensional variables, and define each variable as a characteristic index, calculate the entropy value of each characteristic index, and determine the objective weight of each variable according to the entropy value of each characteristic index. The method for obtaining the objective weight of each variable is as follows:

[0075]

[0076] In the formula, H i is the entropy value of the i-th feature index, n is the total dimension of the time series variable of each scene, T is the total number of time series scene objects, and f ij is the proportion of the jth temporal scene object at the i-th feature index, s ij is the data of the variable, p i is the entropy weight of the ith feature index, N WD 、N Load 、N PV are the dimensional variables of each time series scenario.

[0077] In the specific implementation, the data is compressed to the interval [0,1] by the following formula:

[0078]

[0079] Where P ij is the wind and solar power output or load of the i-th node in the j-th scenario.

[0080] When the variable volatility is small and H i When it is larger, its clustering effect is smaller and its contribution to the selection of extreme scenarios is smaller. i The smaller it is, the greater the clustering influence of the variable is, and the greater its contribution to the selection of extreme scenarios. When all the data of a variable in a scenario are equal, H i =H max =1, its clustering influence is zero.

[0081] In step S211, the subjective weight of each variable is calculated according to the capacity of each wind-solar-load variable; that is, the wind power and load capacity ratio is used as an alternative to subjective weighting.

[0082] In step S212, the objective weights and subjective weights of the variables are weighted averaged to obtain the combined weights of the variables. The specific method of using the combined weights as the weights of the wind-solar-load variables is:

[0083]

[0084] In the formula, w i is the combined weight of each variable, p i is the objective weight of the i-th characteristic index determined by the entropy weight method, q i is the subjective weight of the determined i-th characteristic index.

[0085] S22. Clustering the time series scenarios based on weighted Euclidean distance according to the weights of the wind-solar-load variables to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics.

[0086] Specifically, step S22 adopts a weighted clustering method to perform weighted clustering on time series scenarios in a time dimension to obtain a plurality of typical wind-solar-load operation scenarios.

[0087] In this embodiment, the K-means clustering algorithm is used to pre-assign data objects to the nearest class based on the principle of minimizing the distance to the selected k initial cluster centers, and the data set is divided into different categories during the iteration process to optimize the criterion function for evaluating clustering performance.

[0088] Specifically, according to the time series scene matrix data set x = {x i |x i ∈R h ,i=1,2,...,t}, where h is the dimension of the data variable, i is the number of data in the data set, and k initial clustering centers c are selected. 1 ,c 2 ,…,c k ;

[0089] According to the principle that the distance between each data object and the selected k initial cluster centers is the smallest, the data object is pre-assigned to the nearest class. The weighted Euclidean distance is calculated as follows:

[0090]

[0091] In the formula, x i is the i-th element in a given data set, x j is the jth element in a given data set, d(x i ,x j ) is x i and x j The weighted Euclidean distance between v is the weight of the v-th dimension variable of the time series scene data. In this specific implementation, w v It is usually set to 1, h is the dimension of the time series scene data, x iv is the value of the ith data on the vth variable, x jv is the value of the jth data on the vth variable;

[0092] Recalculate the cluster centers, defined as follows:

[0093]

[0094] Where N j It is a class C j The number of samples in C 1 ,C 2 ,…,C k is the cluster center c 1 ,c 2,…,c k The data set of the class to which it belongs;

[0095] The clustering problem is converted into an optimization problem to obtain the optimal clustering result. The calculation formula of the objective function is as follows:

[0096]

[0097] S3. Based on each typical wind-solar-load operation scenario, the time sequence scenario farthest from the cluster center corresponding to the typical wind-solar-load operation scenario is used as the extreme scenario corresponding to the typical wind-solar-load operation scenario; that is:

[0098] S31. Extract edge running points far away from the cluster center as extreme scenarios.

[0099] S31 specifically includes: based on the wind-solar-load time series scenario, clustering generates several typical wind-solar-load operation scenarios in the time dimension as cluster centers. In each cluster, the edge points far from the cluster center are taken as the extreme values ​​corresponding to each typical wind-solar-load operation scenario, and multiple typical wind-solar-load operation scenarios correspond to multiple extreme scenarios. The extreme scenario extraction method is as follows: Figure 3 As shown, the extreme operating scenario is represented as:

[0100]

[0101] Where i is the number of classes, e i is the edge point associated with the i-th class.

[0102] When n edge points are taken from each category, that is, each typical scenario corresponds to n extreme scenarios, there will be a total of n×k extreme operation scenarios, which are defined as:

[0103]

[0104] Where m represents the extreme scenario of the mth layer corresponding to each typical scenario, m = 1, 2, ..., n, e i m is the mth level extreme scenario corresponding to the ith typical scenario.

[0105] The edge points of the first layer represent the worst extreme cases in each category. Due to the clustering method, the extreme scenarios depend largely on the typical scenarios. These extreme scenarios may cover extreme cases of wind and load, or even uniform cases, which means that the extreme scenarios are not only located outside the contours of the sample data points, but may also be located in the area surrounded by the data points, depending on the number of typical scenarios selected.

[0106] In order to improve the efficiency of the extraction method, this embodiment uses a clustering evaluation index to determine the optimal number of clusters, that is, by calculating the sum of the weighted Euclidean distances between the extreme points and their corresponding cluster centers. We call this sum the extreme point square error, which is expressed as:

[0107]

[0108] In the formula, J e is the clustering evaluation index, E i is the set of extreme points corresponding to the i-th cluster center, and k is the number of clusters.

[0109] In a second aspect, the present invention provides a multi-time scale wind-solar-load extreme scenario extraction system, characterized in that it comprises: a data acquisition module, a typical scenario module and an extreme scenario module;

[0110] The data acquisition module is used to acquire the hourly time series data of wind, light and load of each node in the power grid in a normal year as scene data to form a time series scene;

[0111] The typical scenario module is used to cluster the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics;

[0112] The extreme scenario module is used to take the time sequence scenario farthest from the cluster center corresponding to each typical wind-solar-load operation scenario as the extreme scenario corresponding to the typical wind-solar-load operation scenario based on each typical wind-solar-load operation scenario.

[0113] In a specific implementation, the typical scenario module includes: an empowerment unit and a typical scenario unit.

[0114] The weighting unit in the typical scenario module is used to weight the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario. The wind-solar-load variables include wind power, photovoltaic power and load of each node in the power grid.

[0115] In a specific implementation, the weighting unit includes: an objective weighting subunit, a subjective weighting subunit and a combined weighting subunit;

[0116] The objective weighting subunit is used to weight the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario by using the entropy weight method to obtain the objective weight of each variable, wherein the wind-solar-load variable includes wind power, photovoltaic power and load of each node in the power grid;

[0117] The subjective weighting subunit is used to calculate the subjective weight of each variable according to the capacity of each wind-solar-load variable;

[0118] The combined weighting subunit is used to perform weighted averaging on the objective weighting subunit and the subjective weighting subunit to obtain the combined weight of each variable, and use the combined weight as the weight of each wind-solar-load variable.

[0119] In this embodiment, the types, dimensional differences and importance of clustering variables are significant, and clustering weights need to be set to reflect the characteristics of each variable. The traditional Euclidean distance fails to fully reflect the importance of variables in each dimension. For example, the contribution of wind and solar power generation variables in different locations to extreme scenarios will vary depending on their impact on the research problem. Therefore, in the clustering process, the importance of variables is different, and subjective weights and objective weights need to be considered. In terms of objective weights, variables with large time fluctuations often have a greater impact on the extraction of extreme scenes. The greater the variable fluctuation, the farther the data edge point is from the cluster center point, and the higher the possibility of contributing to the extreme edge point. In terms of subjective weights, it is necessary to analyze the impact of each variable on the research problem and give larger weights to variables with greater influence.

[0120] This embodiment uses the entropy weight method to assign weights to the wind-solar-load variables to obtain the objective weights of each variable. The basic principle is to measure the differences between variables by entropy values ​​and set the weights of each variable. It only reflects the volatility of variable data and has nothing to do with the type and relationship of the variables. According to the characteristics of the entropy weight method, entropy can be used to determine the degree of discreteness of variables. Based on this principle, calculating the Euclidean distance between scenes with entropy weight variables can obtain more accurate results.

[0121] Specifically, in the objective weighting subunit, the entropy weight method is used to determine the wind-solar-load variable weights according to the time series data of each wind-solar-load variable in the time series scenario, and the specific method of obtaining the objective weights of each variable is as follows:

[0122] Assume that there are T time series scene objects to be clustered in the time dimension, and each of the time series scenes contains N WD +N Load +N PV dimensional variables, and define each variable as a characteristic index, calculate the entropy value of each characteristic index, and determine the objective weight of each variable according to the entropy value of each characteristic index. The method for obtaining the objective weight of each variable is as follows:

[0123]

[0124] In the formula, H i is the entropy value of the i-th feature index, n is the total dimension of the time series variable of each scene, T is the total number of time series scene objects, and f ij is the proportion of the jth temporal scene object at the i-th feature index, s ij is the data of the variable, p i is the entropy weight of the ith feature index, NWD 、N Load 、N PV are the dimensional variables of each time series scenario.

[0125] In the specific implementation, the data is compressed to the interval [0,1] by the following formula:

[0126]

[0127] Where P ij is the wind and solar power output or load of the i-th node in the j-th scenario.

[0128] When the variable volatility is small and H i When it is larger, its clustering effect is smaller and its contribution to the selection of extreme scenarios is smaller. i The smaller it is, the greater the clustering influence of the variable is, and the greater its contribution to the selection of extreme scenarios. When all the data of a variable in a scenario are equal, H i =H max =1, its clustering influence is zero.

[0129] Specifically, in the subjective weighting subunit, the subjective weight of each variable is calculated according to the capacity capability of each wind-solar-load variable, that is, the wind power and load capacity ratio is used as an alternative to subjective weighting.

[0130] Specifically, in the combined weighting subunit, the objective weight and subjective weight of each variable are weighted averaged to obtain the combined weight of each variable. The specific method of using the combined weight as the weight of each wind-solar-load variable is:

[0131]

[0132] In the formula, w i is the combined weight of each variable, p i is the objective weight of the i-th characteristic index determined by the entropy weight method, q i is the subjective weight of the determined i-th characteristic index.

[0133] In a specific implementation, the typical scenario unit in the typical scenario module is used to cluster the time series scenarios based on the weighted Euclidean distance according to the weights of the wind-solar-load variables to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics.

[0134] Specifically, in the typical scenario unit, the weighted clustering method is adopted to obtain multiple typical wind-solar-load operation scenarios by weighted clustering the time series scenarios in the time dimension.

[0135] In this embodiment, the K-means clustering algorithm is used to pre-assign data objects to the nearest class based on the principle of minimizing the distance to the selected k initial cluster centers, and the data set is divided into different categories during the iteration process to optimize the criterion function for evaluating clustering performance.

[0136] Specifically, according to the time series scene matrix data set x = {x i |x i ∈R h ,i=1,2,...,t}, where h is the dimension of the data variable, i is the number of data in the data set, and k initial clustering centers c are selected. 1 ,c 2 ,…,c k ;

[0137] According to the principle that the distance between each data object and the selected k initial cluster centers is the smallest, the data object is pre-assigned to the nearest class. The weighted Euclidean distance is calculated as follows:

[0138]

[0139] In the formula, x i is the i-th element in a given data set, x j is the jth element in a given data set, d(x i ,x j ) is x i and x j The weighted Euclidean distance between v is the weight of the v-th dimension variable of the time series scene data. In this specific implementation, w v It is usually set to 1, h is the dimension of the time series scene data, x iv is the value of the ith data on the vth variable, x jv is the value of the jth data on the vth variable;

[0140] Recalculate the cluster centers, defined as follows:

[0141]

[0142] Where N j It is a class C j The number of samples in C 1 ,C 2 ,…,C k is the cluster center c 1 ,c 2 ,…,c k The data set of the class to which it belongs;

[0143] The clustering problem is converted into an optimization problem to obtain the optimal clustering result. The calculation formula of the objective function is as follows:

[0144]

[0145] In the specific implementation, the extreme scenario module is specifically based on the wind-solar-load timing scenario, and clusters and generates several typical wind-solar-load operation scenarios in the time dimension as clustering centers.

[0146] In each cluster, the edge points far from the cluster center are taken as the extreme values ​​corresponding to each typical wind-solar-load operation scenario. Multiple typical wind-solar-load operation scenarios correspond to multiple extreme scenarios. The extreme scenario extraction method is as follows: Figure 3 As shown, the extreme operating scenario is represented as:

[0147]

[0148] Where i is the number of classes, e i is the edge point associated with the i-th class.

[0149] When n edge points are taken from each category, that is, each typical scenario corresponds to n extreme scenarios, there will be a total of n×k extreme operation scenarios, which are defined as:

[0150]

[0151] Where m represents the extreme scenario of the mth layer corresponding to each typical scenario, m = 1, 2, ..., n, is the mth level extreme scenario corresponding to the ith typical scenario.

[0152] The edge points of the first layer represent the worst extreme cases in each category. Due to the clustering method, the extreme scenarios depend largely on the typical scenarios. These extreme scenarios may cover extreme cases of wind and load, or even uniform cases, which means that the extreme scenarios are not only located outside the contours of the sample data points, but may also be located in the area surrounded by the data points, depending on the number of typical scenarios selected.

[0153] In order to improve the efficiency of the extraction method, this embodiment uses a clustering evaluation index to determine the optimal number of clusters, that is, by calculating the sum of the weighted Euclidean distances between the extreme points and their corresponding cluster centers. We call this sum the extreme point square error, which is expressed as:

[0154]

[0155] In the formula, J e is the clustering evaluation index, E i is the set of extreme points corresponding to the i-th cluster center, and k is the number of clusters.

[0156] In a third aspect, the present invention also provides a storage medium comprising a computer executable program, which, when executed by a computer processor, is used to execute the multi-time-scale wind-light-load extreme scene extraction method of the present invention.

[0157] Among them, the storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. The computer-readable storage medium may include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device or device.

[0158] The code of the computer executable program for performing the operation of the present invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, ++, and also conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network or a wide area network, or can be connected to an external computer.

[0159] In order to illustrate the effectiveness of this method more clearly and intuitively, the following verification description is given for this method.

[0160] A wind farm, a photovoltaic farm and a node load are selected as the main variables. The capacities of the wind farm and photovoltaic farm are 100MW and 50MW respectively, and the maximum load of the node is 150MW.

[0161] In traditional transmission network planning, the annual maximum load scenario is usually taken as the scenario with the highest risk to verify the safety of the planning scheme. This is usually the operating scenario with the greatest network pressure. However, with the grid connection of large-scale intermittent renewable energy, the annual maximum load scenario is no longer the worst scenario for the planning scheme. Figure 4As shown in Figure 2, the worst extreme wind-solar-load scenario extracted by the clustering method is no longer the scenario with the maximum or minimum load, but occurs at 4618 hours and 8151 hours. The grid connection of wind power has led to changes in the system's extreme operating scenarios.

[0162] This model uses the entropy weight method and the capacity method to determine the objective and subjective weight ratios of the variables and combine them. In Table 1, the original weights of the three variables are set to 0.328. Although load power has a higher value and has the largest subjective weight among the three variables, its volatility is still smaller than that of wind and solar output. This shows that load power has a weaker impact on extreme scenarios, so its cluster combination weight is still small.

[0163] Table 1 Clustering variable weights

[0164]

[0165] In order to construct the wind-solar-load scenario, the wind farm and the photovoltaic farm are considered as one variable, regardless of the location of the wind and photovoltaic. By taking the number of clusters as 3, the worst extreme scenarios numbered [4618, 8759, 5656] are obtained; by considering their different geographical locations, the wind and photovoltaic are assumed to be separate cluster variables, and the extreme scenarios obtained by the clustering method are numbered [4618, 8151, 8656]. The extreme situations have changed, and the corresponding wind-solar-load situations are shown in Table 2.

[0166] Table 2 Comparison of extreme scenarios with distributed clustering and centralized clustering

[0167]

[0168] Depend on Figure 5 It can be seen that considering the geographical distribution of wind and solar power, the wind-solar-load operation scenario obtained by clustering is more extreme at the global level than that obtained by centralized clustering. This scenario also has a higher load level and a lower wind-solar output level. Therefore, the different geographical locations of wind and solar power generation must be considered in the clustering method to extract more extreme scenarios.

[0169] As the number of clusters increases, a larger number of extreme scenarios with more comprehensive information can be derived. The growth in the number of clusters increases the computational requirements and leads to the extraction of redundant extreme scenarios. Figure 6 It can be seen that for cluster numbers above 3, the EPSE standard does not change much, that is, increasing the number of clusters does not produce positive feedback.

[0170] In summary, the present invention proposes a method for extracting extreme operation scenarios of wind-solar-load based on the idea of ​​weighted clustering. Under this method, the wind-solar-load time series is selected as the clustering object, and the operating point farthest from the weighted Euclidean center of the cluster is identified as an extreme scenario. With this contribution, large-scale wind and solar can be found in the planning stage and stability assessment and flow analysis can be performed on the weak links of power grid operation. The results prove the effectiveness of the new method. It can enable planners to quickly evaluate the operational safety of system planning schemes and improve the rationality and scientificity of large-scale power grid stability scenario analysis.

[0171] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described with reference to the preferred embodiments of the present invention, it should be understood by those skilled in the art that various changes may be made in form and details without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A multi-time scale wind-solar-load extreme scene extraction method, characterized in that: The steps include: Obtain the hourly time series data of wind, light and load of each node in the power grid in the normal year as scenario data to form a time series scenario; Clustering the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics; Based on each typical wind-solar-load operation scenario, the time sequence scenario corresponding to the typical wind-solar-load operation scenario and farthest from the cluster center is used as the extreme scenario corresponding to the typical wind-solar-load operation scenario.

2. The multi-timescale wind-solar-load extreme scene extraction method according to claim 1 is characterized in that: The time series scenarios are clustered in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on the combination of time correlation and geographical distribution characteristics, including: According to the time series data of each wind-solar-load variable in the time series scenario, weight the wind-solar-load variable, wherein the wind-solar-load variable includes wind power, photovoltaic power and load of each node in the power grid; According to the weights of the wind-solar-load variables, the time series scenarios are clustered based on weighted Euclidean distance to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics.

3. The multi-timescale wind-solar-load extreme scene extraction method according to claim 2 is characterized in that: The weighting of the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario specifically includes: According to the time series data of each wind-solar-load variable in the time series scenario, the wind-solar-load variable is weighted by using the entropy weight method to obtain the objective weight of each variable; According to the capacity of each wind-solar-load variable, the subjective weight of each variable is calculated; The objective weights and subjective weights of the variables are weighted averaged to obtain the combined weights of the variables, and the combined weights are used as the weights of the wind-solar-load variables.

4. The multi-timescale wind-solar-load extreme scene extraction method according to claim 3 is characterized in that: The specific method of determining the wind-solar-load variable weights by using the entropy weight method according to the time series data of each wind-solar-load variable in the time series scenario and obtaining the objective weights of each variable is as follows: Assume that there are T time series scene objects to be clustered in the time dimension, and each of the time series scenes contains N WD +N Load +N PV dimensional variables, and define each variable as a characteristic index, calculate the entropy value of each characteristic index, and determine the objective weight of each variable according to the entropy value of each characteristic index. The method for obtaining the objective weight of each variable is as follows: In the formula, H i is the entropy value of the i-th feature index, n is the total dimension of the time series variable of each scene, T is the total number of time series scene objects, and f ij is the proportion of the jth temporal scene object at the i-th feature index, s ij is the data of the variable, p i is the entropy weight of the i-th feature index, N WD 、N Load 、N PV are the dimensional variables of each time series scenario.

5. The multi-timescale wind-solar-load extreme scene extraction method according to claim 3 is characterized in that: The specific method of performing weighted average of the objective weight and subjective weight of each variable to obtain the combined weight of each variable is: In the formula, w i is the combined weight of each variable, p i is the objective weight of the i-th characteristic index determined by the entropy weight method, q i is the subjective weight of the determined i-th characteristic index.

6. The multi-timescale wind-solar-load extreme scene extraction method according to claim 2 is characterized in that: The calculation formula of the weighted Euclidean distance is as follows: In the formula, x i is the i-th element in the time series scene matrix dataset, x j is the jth element in the time series scene dataset, d(x i ,x j ) is x i and x j The weighted Euclidean distance between v is the weight of the v-th dimension variable of the time series scene data, h is the dimension of the time series scene data, x iv is the value of the ith data on the vth variable, x jv is the value of the j-th data on the v-th variable.

7. The multi-timescale wind-solar-load extreme scene extraction method according to claim 1 is characterized in that: Based on each typical wind-solar-load operation scenario, the time sequence scenario corresponding to the typical wind-solar-load operation scenario and the one farthest from the cluster center as the extreme scenario corresponding to the typical wind-solar-load operation scenario specifically includes: For each typical wind-solar-load operation scenario, according to the preset required number, from the class corresponding to the typical wind-solar-load operation scenario, among the time series scenes with the farthest weighted Euclidean distance, a preset number of time series scenes farthest from the cluster center are selected as the extreme scenes corresponding to the typical wind-solar-load operation scenario.

8. A multi-time scale wind-solar-load extreme scene extraction system, characterized in that: include: Data acquisition module, typical scenario module and extreme scenario module; The data acquisition module is used to acquire the hourly time series data of wind, light and load of each node in the power grid in a normal year as scene data to form a time series scene; The typical scenario module is used to cluster the time series scenarios in the time dimension to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics; The extreme scenario module is used to take the time sequence scenario farthest from the cluster center corresponding to each typical wind-solar-load operation scenario as the extreme scenario corresponding to the typical wind-solar-load operation scenario based on each typical wind-solar-load operation scenario.

9. The multi-timescale wind-solar-load extreme scene extraction system according to claim 8, characterized in that: The typical scene module includes: a weighting unit and a typical scene unit; The weighting unit is used to weight the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario, and the wind-solar-load variables include wind power, photovoltaic power and load of each node in the power grid; The typical scenario unit is used to cluster the time series scenarios based on the weighted Euclidean distance according to the weights of the wind-solar-load variables, so as to obtain multiple typical wind-solar-load operation scenarios based on a combination of time correlation and geographical distribution characteristics.

10. The multi-time scale wind-solar-load extreme scene extraction system according to claim 8, characterized in that: The weighting unit includes: an objective weighting subunit, a subjective weighting subunit and a combined weighting subunit; The objective weighting subunit is used to weight the wind-solar-load variables according to the time series data of each wind-solar-load variable in the time series scenario by using the entropy weight method to obtain the objective weight of each variable, wherein the wind-solar-load variable includes wind power, photovoltaic power and load of each node in the power grid; The subjective weighting subunit is used to calculate the subjective weight of each variable according to the capacity of each wind-solar-load variable; The combined weighting subunit is used to perform weighted averaging on the objective weighting subunit and the subjective weighting subunit to obtain the combined weight of each variable.

11. A storage medium containing a computer executable program, characterized in that: When the computer executable program is executed by a computer processor, it is used to execute the multi-time scale wind-light-load extreme scene extraction method according to any one of claims 1 to 7.

Images