Method, device and equipment for calculating credibility of wind and light capacity in consideration of extreme climate scene and storage medium

Abstract

The invention relates to the technical field of electric power, and provides a calculation method, device and equipment for wind and light capacity credibility in consideration of an extreme climate scene and a storage medium, and the method comprises the steps: obtaining historical climate data, multi-dimensional wind and light load historical data, and power grid basic data; determining an extreme climate scene and a typical climate scene according to the historical climate data; constructing a load condition probability distribution model according to the multi-dimensional wind and light load historical data; and according to the multi-dimensional wind-light load historical data, the power grid basic data and the load condition probability distribution model, calculating the wind-light capacity credibility in an extreme climate scene and the wind-light capacity credibility in a typical climate scene. According to the method provided by the invention, the problem that the existing capacity credibility calculation does not involve an extreme climate scene is solved; when the day-ahead power generation plan of the system is arranged, the influence of climate conditions on the credibility of the wind-light capacity is comprehensively considered, and the power generation plan arrangement of a conventional unit can be more reasonable in combination with wind-light processing prediction.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and specifically to a calculation method, device, equipment and storage medium for the credibility of wind and solar power capacity considering extreme climate scenarios. Background Art

[0002] In recent years, the installed capacity of grid-connected photovoltaic power generation and wind power generation has grown rapidly. The introduction of power sources with uncertain, intermittent and uncontrollable output on the power generation side of the power system has brought challenges to the power system in maintaining the real-time balance of power generation and load.

[0003] The credibility of wind and solar capacity enables the installed capacity of intermittent wind and photovoltaic generators to be compared with that of conventional generators on the same level. It makes it possible to analyze and calculate the uncontrollable, volatile and random wind and solar power generation in combination with the traditional power system. This is of great significance to power balance, power reserve rate calculation, capacity market pricing, etc. in power planning.

[0004] However, at present, there is still a lack of research on considering the impact of extreme climate scenarios in the credibility assessment of wind and solar capacity. The credibility assessment of wind and solar capacity in extreme climate scenarios has become a difficult problem. Summary of the invention

[0005] To solve the above problems, the present application provides a calculation method, device, equipment and storage medium for the credibility of wind and solar capacity under extreme climate scenarios, extracts extreme climate scenarios, and thus realizes the credibility assessment of wind and solar capacity under extreme climate scenarios.

[0006] This application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a method for calculating the credibility of wind and solar capacity considering extreme climate scenarios, the method comprising:

[0008] Obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data;

[0009] Determine extreme climate scenarios and typical climate scenarios based on historical climate data;

[0010] Construct a load condition probability distribution model based on multi-dimensional wind and solar load historical data;

[0011] Based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model, the credibility of wind and solar capacity under extreme climate scenarios and the credibility of wind and solar capacity under typical climate scenarios are calculated.

[0012] Optionally, extreme climate scenarios and typical climate scenarios are determined based on historical climate data, including:

[0013] The elbow method and k-means++ algorithm were used to cluster the historical climate data;

[0014] The extreme climate scenarios and typical climate scenarios are determined based on the distribution of historical climate data in the clusters.

[0015] Optionally, the elbow method and k-means++ algorithm are used to cluster the historical climate data, including:

[0016] The elbow method is used to determine the target number of clusters for clustering;

[0017] The k-means++ algorithm is used to determine the initial centroid of each cluster;

[0018] Calculate the Euclidean distance from each historical climate data to each initial centroid, classify each historical climate data into the initial centroid with the smallest Euclidean distance, recalculate the centroid for each cluster using the historical climate data classified into the cluster, and determine the cluster center according to the final centroid of each cluster after iteration until the stopping condition is reached;

[0019] According to the distribution of historical climate data in the cluster, extreme climate scenarios and typical climate scenarios are determined, including:

[0020] Calculate the Euclidean distance from each historical climate data to the cluster center of the cluster to which it belongs;

[0021] For each cluster, select a preset number of historical climate data with the largest Euclidean distance as special climate data;

[0022] Determine the extreme climate scenario based on the date corresponding to the special climate data, and determine the typical climate scenario based on the extreme climate scenario.

[0023] Optionally, a load condition probability distribution model is constructed based on multi-dimensional wind and solar load historical data, including;

[0024] The D-vine structure and non-parametric kernel density estimation method are used to model the correlation of multi-dimensional wind and solar load historical data and obtain the load condition probability distribution model.

[0025] Optionally, the D-vine structure and non-parametric kernel density estimation method are used to model the correlation of multi-dimensional wind and solar load historical data to obtain a load condition probability distribution model, including:

[0026] The non-parametric kernel density estimation method is used to perform probability integral transformation on the historical data of wind and solar loads.

[0027] Based on the sample set after probability integral transformation, the function type and parameters of each pair copula in the D-vine structure are optimized;

[0028] Based on the optimized D-vine structure, a load condition probability distribution model is established with the output of wind turbines and photovoltaic generators as conditional variables.

[0029] Optionally, the credibility of wind and solar capacity under extreme climate scenarios and the credibility of wind and solar capacity under typical climate scenarios are calculated based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model, including:

[0030] The expected value of power shortage is used as a reliability indicator;

[0031] Based on the multi-dimensional wind and solar load historical data and basic grid data, determine the benchmark reliability index of the system at each moment that does not include wind turbines and photovoltaic generators;

[0032] Determine the dynamic reliability index of the system including wind turbines and photovoltaic generators at each moment based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model;

[0033] When the difference between the dynamic reliability index at the corresponding moment and the benchmark reliability index is less than the preset threshold, the adjustment amount of the startup capacity of the conventional unit is determined as the wind and solar operation reliable capacity at each moment;

[0034] According to the dates corresponding to the extreme climate scenarios and the dates corresponding to the typical climate scenarios, based on the wind and solar credible capacity at each moment, the wind and solar credible capacity under the extreme climate scenarios and the wind and solar credible capacity under the typical climate scenarios are calculated respectively;

[0035] Based on the wind and solar operation reliable capacity and wind and solar installed capacity, the wind and solar capacity reliability under extreme climate scenarios and the wind and solar capacity reliability under typical climate scenarios are calculated.

[0036] Optionally, based on the multi-dimensional wind and solar load historical data and the basic data of the power grid, the benchmark reliability index of the system at each moment excluding the wind turbine generator set and the photovoltaic generator set is determined, including:

[0037] Determine the operating capacity of conventional units based on the installed capacity of conventional units, forced outage rate, peak load, and spinning reserve rate;

[0038] According to the startup capacity of conventional units and the load data at each moment, the Monte Carlo method is used to determine the benchmark reliability index at each moment;

[0039] Based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model, the dynamic reliability indicators of the system including wind turbines and photovoltaic generators at each moment are determined, including:

[0040] Use the binary method or truncation method to adjust the startup capacity of conventional units;

[0041] According to the adjusted startup capacity of conventional units, the probability distribution model of load conditions, and the load data at each moment, the Monte Carlo method is used to determine the dynamic reliability index at each moment.

[0042] In a second aspect, the present application provides a computing device for the credibility of wind and solar power capacity considering extreme climate scenarios, the device comprising:

[0043] Data unit, used to obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data;

[0044] A scenario extraction unit, used to determine extreme climate scenarios and typical climate scenarios based on historical climate data;

[0045] A model building unit, used to build a load condition probability distribution model based on multi-dimensional wind and solar load historical data;

[0046] The solving unit is used to calculate the credibility of wind and solar capacity under extreme climate scenarios and the credibility of wind and solar capacity under typical climate scenarios based on multi-dimensional wind and solar load historical data, basic power grid data, and load condition probability distribution model.

[0047] Optionally, a scene extraction unit comprises:

[0048] Clustering module, used to cluster historical climate data using the elbow method and k-means++ algorithm;

[0049] An extraction module is used to determine extreme climate scenarios and typical climate scenarios based on the distribution of historical climate data in clusters;

[0050] Optionally, the clustering module is also used to:

[0051] The elbow method is used to determine the target number of clusters for clustering;

[0052] The k-means++ algorithm is used to determine the initial centroid of each cluster;

[0053] The Euclidean distance from each historical climate data to each initial centroid is calculated, and each historical climate data is classified into the initial centroid with the smallest Euclidean distance. For each cluster, the centroid is recalculated using the historical climate data classified into the cluster. After iterating until the stopping condition is reached, the cluster center is determined according to the final centroid of each cluster.

[0054] Extraction module, also used for:

[0055] Calculate the Euclidean distance from each historical climate data to the cluster center of the cluster to which it belongs;

[0056] For each cluster, select a preset number of historical climate data with the largest Euclidean distance as special climate data;

[0057] Determine the extreme climate scenario based on the date corresponding to the special climate data, and determine the typical climate scenario based on the extreme climate scenario.

[0058] Optionally, the model building unit is used to:

[0059] The D-vine structure and non-parametric kernel density estimation method are used to model the correlation of multi-dimensional wind and solar load historical data and obtain the load condition probability distribution model.

[0060] Optionally, the model building unit is also used to:

[0061] The non-parametric kernel density estimation method is used to perform probability integral transformation on the historical data of wind and solar loads.

[0062] Based on the sample set after probability integral transformation, the function type and parameters of each pair copula in the D-vine structure are optimized;

[0063] Based on the optimized D-vine structure, a load condition probability distribution model is established with the output of wind turbines and photovoltaic generators as conditional variables.

[0064] Optional, solver unit, including:

[0065] A benchmark reliability index determination module is used to determine the benchmark reliability index of the system excluding wind turbines and photovoltaic generators at each moment according to the multi-dimensional wind, solar and load historical data and the basic data of the power grid, wherein the power shortage expected value is used as the reliability index;

[0066] A dynamic reliability index determination module is used to determine the dynamic reliability index of the system including the wind turbine generator set and the photovoltaic generator set at each moment according to the multi-dimensional wind and solar load historical data, the basic data of the power grid, and the load condition probability distribution model;

[0067] The wind-solar operation credible capacity calculation module is used to determine the adjustment amount of the conventional unit startup capacity as the wind-solar operation credible capacity at each moment when the difference between the dynamic reliability index at the corresponding moment and the benchmark reliability index is less than a preset threshold value, and calculate the wind-solar operation credible capacity under the extreme climate scenario and the wind-solar operation credible capacity under the typical climate scenario based on the wind-solar operation credible capacity at each moment according to the date corresponding to the extreme climate scenario and the date corresponding to the typical climate scenario;

[0068] The wind and solar capacity credibility calculation module is used to calculate the wind and solar capacity credibility under extreme climate scenarios and the wind and solar capacity credibility under typical climate scenarios based on the wind and solar operation credibility capacity and the wind and solar installed capacity.

[0069] Optionally, the benchmark reliability index determination module is also used to:

[0070] Determine the operating capacity of conventional units based on the installed capacity of conventional units, forced outage rate, peak load, and spinning reserve rate;

[0071] According to the startup capacity of conventional units and the load data at each moment, the Monte Carlo method is used to determine the benchmark reliability index at each moment;

[0072] The dynamic reliability index determination module is also used to: adjust the startup capacity of conventional units by using the dichotomy method or the truncation method;

[0073] According to the adjusted startup capacity of conventional units, the probability distribution model of load conditions, and the load data at each moment, the Monte Carlo method is used to determine the dynamic reliability index at each moment.

[0074] In a third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for calculating the credibility of wind and solar capacity considering extreme climate scenarios are implemented.

[0075] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is instructed by a processor, it implements the steps of the above-mentioned method for calculating the credibility of wind and solar capacity under extreme climate scenarios.

[0076] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0077] The calculation method of wind and solar capacity credibility in extreme climate scenarios provided by the present application first obtains historical climate data, multi-dimensional wind and solar load historical data, and basic grid data; then determines extreme climate scenarios and typical climate scenarios based on historical climate data; then builds a load condition probability distribution model based on multi-dimensional wind and solar load historical data; thereby calculating the wind and solar capacity credibility in extreme climate scenarios and the wind and solar capacity credibility in typical climate scenarios based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model. The calculation method of wind and solar capacity credibility in extreme climate scenarios provided by the present application solves the problem that the existing capacity credibility calculation does not involve extreme climate scenarios; when arranging the system's day-ahead power generation plan, comprehensively considering the impact of climate conditions on wind and solar capacity credibility, and combining with wind and solar processing predictions, the power generation plan arrangement of conventional units can be made more reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0079] Figure 1 A schematic flow chart showing a method for calculating the credibility of wind and solar capacity in extreme climate scenarios according to an embodiment of the present application is shown;

[0080] Figure 2 A schematic diagram showing a process of determining extreme climate scenarios and typical climate scenarios based on historical climate data according to an embodiment of the present application;

[0081] Figure 3 A schematic diagram showing the output of a wind turbine generator set and the output of a photovoltaic generator set on June 24, 2023 according to an embodiment of the present application;

[0082] Figure 4 A schematic diagram of a process for calculating the wind-solar capacity credibility under extreme climate scenarios and the wind-solar capacity credibility under typical climate scenarios according to multi-dimensional wind-solar load history data, basic grid data, and a load condition probability distribution model according to an embodiment of the present application is shown;

[0083] Figure 5 A schematic diagram showing the structure of a device for calculating the credibility of wind and solar capacity in consideration of extreme climate scenarios according to an embodiment of the present application;

[0084] Figure 6 A schematic structural diagram of a computer device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0086] The idea of ​​this application is to propose a method for extracting extreme climate scenarios and calculating the credibility of wind and solar capacity under extreme climate scenarios in view of the drawbacks of existing wind and solar capacity credibility and credible capacity evaluation that mainly focuses on typical scenarios. The application is described in detail below through specific embodiments.

[0087] Figure 1A flow chart of a method for calculating the credibility of wind and solar capacity in extreme climate scenarios proposed in an embodiment of the present application is shown. Figure 1 As shown, this embodiment includes steps S110 to S140:

[0088] Step S110, obtaining historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data.

[0089] Obtain basic data related to power grid operation, which may include historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data.

[0090] Historical climate data can be used to indicate climate characteristic data for each time period of each day in the retrospective period. For example, historical climate data can include climate characteristic data for three hours every day throughout the year of 2022; for another example, historical climate data can include climate characteristic data for one hour every day throughout the year of 2023. Climate characteristic data can include, but are not limited to: temperature, average wind speed, maximum wind speed, precipitation, horizontal visibility, air pressure, etc.

[0091] Multi-dimensional wind, photovoltaic and load historical data can be used to indicate wind power generation data, photovoltaic power generation data, load data, etc. at each time of each day in the retrospective period. For example, multi-dimensional wind, photovoltaic and load historical data can include wind power generation data, photovoltaic power generation data, load data, etc. at each hour of each day in the whole year of 2022; for another example, multi-dimensional wind, photovoltaic and load historical data can include wind power generation data, photovoltaic power generation data, load data, etc. at even hours of each day in the whole year of 2023. Wind power generation data at least includes the active power of wind turbines (i.e., wind turbine output data), and photovoltaic power generation data at least includes the active power of photovoltaic generators (i.e., photovoltaic generator output data).

[0092] Basic data of the power grid can be used to indicate the power supply data of the power system. The power supply data may include, but is not limited to: conventional installed capacity, forced outage rate, peak load, spinning reserve rate, etc. Conventional installed capacity can be used to indicate the actual rated active power of the system. The forced outage rate can be used to indicate the ratio of the forced outage time of the equipment during the calculation period to the length of the calculation period. Spinning reserve can be used to indicate the synchronous capacity that the system can quickly adjust to ensure stable and reliable power supply in the event of sudden failures in the power system or uncertain load growth. Spinning reserve may include, but is not limited to, accident reserve and load reserve. The spinning reserve rate is usually 2% to 5% of the peak load.

[0093] If necessary, the acquired historical climate data, multi-dimensional wind and solar load historical data, and basic grid data can be preprocessed. The preprocessing may include, but is not limited to, time alignment, format conversion, etc. The preprocessed basic data is used for subsequent related calculations.

[0094] Step S120, determining extreme climate scenes and typical climate scenes based on historical climate data.

[0095] The purpose of this application is to calculate the credibility of wind and solar capacity under extreme climate scenarios, so it is necessary to extract extreme climate scenarios based on historical climate data.

[0096] Extreme climate scenarios can be used to indicate that the climate data of a certain date is extremely abnormal and does not conform to the normal meteorological level. For example, extreme climate scenarios can be used to indicate special climate features that differ greatly from the average climate level throughout the year. For example, for areas with relatively stable meteorological levels throughout the year, if the annual climate average level is drought, and the precipitation on a certain date is extremely large, then the date can be extracted as an extreme climate scenario. For another example, extreme climate scenarios can also be used to indicate special climate features that differ greatly from the average climate levels of each season throughout the year. For example, for areas where the annual meteorological level is clearly distinguished by spring, summer, autumn and winter, if the temperature on a certain date is above zero when the winter climate average level is below zero, then the date can be extracted as an extreme climate scenario.

[0097] The method of extracting extreme climate scenes may include, but is not limited to, average comparison, median comparison, clustering, and the like. For example, the average of each climate characteristic data in the historical climate data may be calculated, and the date corresponding to each climate characteristic data with the largest difference from the average value is the extreme climate scene; or the median of each climate characteristic data in the historical climate data may be calculated, and the date corresponding to each climate characteristic data with the largest difference from the median value is the extreme climate scene. Alternatively, the k-means method may be used to cluster the historical climate data, and the date corresponding to the historical climate data with the farthest distance from each cluster center may be determined as the extreme climate scene.

[0098] Step S130, constructing a load condition probability distribution model based on multi-dimensional wind and solar load historical data.

[0099] Since the wind speed, light intensity and load distribution characteristics at different locations at the same time are different, there is a nonlinear correlation between adjacent wind turbines and photovoltaic generators (neither completely independent nor completely correlated), and there is also a nonlinear correlation between wind, light and load. Therefore, in order to calculate the credibility of wind and light capacity, it is necessary to construct a load condition probability distribution model that can reflect the correlation between wind turbine output, photovoltaic generator output and load.

[0100] As a method to study the interdependence between variables, Copula theory connects the joint distribution function with the marginal distribution function of each variable, which can capture the nonlinear relationship and tail correlation between variables and is more suitable for complex data analysis. Therefore, when constructing a load condition probability distribution model, the R-vine structure based on Copula theory, or two special forms of the R-vine structure, the C-vine structure and the D-vine structure, can be used.

[0101] Step S140, calculating the wind-solar capacity credibility under extreme climate scenarios and the wind-solar capacity credibility under typical climate scenarios based on multi-dimensional wind-solar load historical data, basic grid data, and load condition probability distribution model.

[0102] The existing definition of wind and solar capacity credibility is mainly the ratio of equivalent reliable capacity (Equivalent Firm Capacity, EFC). Under this definition, wind and solar capacity credibility is defined as the capacity ratio that can replace 100% reliable conventional units, that is, for the actual system after removing wind and solar units and adding a certain capacity of virtual conventional units with no outage rate, the reliability of the system is the same as the reliability of the actual system. At this time, the ratio of the capacity of virtual conventional units to the installed capacity of wind and solar units is defined as the wind and solar capacity credibility. In other words, the ratio of wind and solar operating credible capacity to wind and solar installed capacity is the wind and solar capacity credibility.

[0103] When calculating the reliable capacity of wind and solar operation, the benchmark reliability index of the system excluding wind turbines and photovoltaic generators at each moment can be determined based on the multi-dimensional wind and solar load historical data and the basic data of the power grid. The dynamic reliability index of the system including wind turbines and photovoltaic generators at each moment can be determined based on the multi-dimensional wind and solar load historical data, the basic data of the power grid and the probability distribution model of load conditions. Then, according to the gap between the benchmark reliability index and the dynamic reliability index at the corresponding moment, the adjustment amount of the startup capacity of the conventional unit can be determined, thereby determining the reliable capacity of wind and solar operation. Reliability indicators can include but are not limited to expected value of insufficient electricity (EENS), loss of load probability (LOLP), power shortage frequency (LOLF), power shortage duration (LOLD), etc. The calculation method of reliability can include but is not limited to: convolution method, Monte Carlo method, intelligent algorithm, etc.

[0104] After calculating the wind and solar power reliable capacity at each moment, the wind and solar power reliable capacity under extreme climate scenarios and the wind and solar power reliable capacity under typical climate scenarios can be calculated based on the extracted dates corresponding to extreme climate scenarios and the dates corresponding to typical climate scenarios. Based on the ratio of wind and solar power reliable capacity to wind and solar power installed capacity, the wind and solar power capacity reliability under extreme climate scenarios and the wind and solar power capacity reliability under typical climate scenarios can be determined.

[0105] The wind and solar capacity credibility under extreme climate scenarios can be expressed as the wind and solar capacity credibility corresponding to the date corresponding to the extreme climate scenario. The wind and solar capacity credibility under typical climate scenarios can be expressed as the wind and solar capacity credibility of other dates after removing the dates corresponding to extreme climate scenarios in the retrospective period; the dates corresponding to extreme climate scenarios can also be removed in the retrospective period, and other dates can be divided according to clusters to determine the wind and solar capacity credibility under the typical climate scenario corresponding to each cluster; the dates corresponding to extreme climate scenarios can also be removed in the retrospective period, and other dates can be divided according to seasons to determine the wind and solar capacity credibility under the typical climate scenario corresponding to each season.

[0106] from Figure 1 It can be seen from the method shown that the calculation method of wind and solar capacity credibility under extreme climate scenarios provided by the present application first obtains historical climate data, multi-dimensional wind and solar load historical data, and basic grid data; then determines extreme climate scenarios and typical climate scenarios based on historical climate data; then builds a load condition probability distribution model based on multi-dimensional wind and solar load historical data; thus, the wind and solar capacity credibility under extreme climate scenarios and the wind and solar capacity credibility under typical climate scenarios are calculated based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model. The calculation method of wind and solar capacity credibility under extreme climate scenarios provided by the present application solves the problem that the existing capacity credibility calculation does not involve extreme climate scenarios; when arranging the system's day-ahead power generation plan, the impact of climate conditions on wind and solar capacity credibility is comprehensively considered, and combined with wind and solar processing predictions, the power generation plan arrangement of conventional units can be made more reasonable.

[0107] In some embodiments of the present application, in the above method, step S120, determining extreme climate scenarios and typical climate scenarios based on historical climate data, includes: clustering the historical climate data using the elbow method and the k-means++ algorithm; determining extreme climate scenarios and typical climate scenarios based on the distribution of the historical climate data in the clusters.

[0108] In this embodiment, a clustering method is used to determine extreme climate scenes and typical climate scenes. K-means is a simple and commonly used clustering method. However, K-means is sensitive to the number of clusters (i.e., the k value), and the selection of the k value will affect the clustering effect; and K-means is sensitive to the centroid of the initial cluster, and the selection of the centroid of the initial cluster will also affect the clustering effect. Therefore, K-means is improved.

[0109] The elbow method is a heuristic method for determining the number of clusters for the optimal clustering in a clustering algorithm. The core of the elbow method is: as the number of clusters increases, the sample division will be more refined, the degree of aggregation of each cluster will gradually increase, and the sum of squared errors (SSE) will gradually decrease. When the k value is less than the number of clusters for the optimal clustering, the SSE will decrease significantly; when the k value reaches the number of clusters for the optimal clustering, the SSE will decrease sharply, and then it will level off as the k value continues to increase. The relationship between SSE and k value is similar to the shape of an elbow, and the k value corresponding to the elbow is the number of clusters for the optimal clustering.

[0110] The k-means++ algorithm can solve the problem of how to select the initial centroid during the clustering process. Compared with the k-means algorithm, which randomly selects the initial centroid, the k-means++ algorithm introduces probability to improve the quality of the solution and shorten the running time.

[0111] Specifically, clustering the historical climate data using the elbow method and the k-means++ algorithm includes: using the elbow method to determine the target number of clusters for clustering; using the k-means++ algorithm to determine the initial centroid of each cluster; calculating the Euclidean distance from each historical climate data to each initial centroid, classifying each historical climate data into the initial centroid with the smallest Euclidean distance, recalculating the centroid for each cluster using the historical climate data classified into the cluster, iterating until the stopping condition is reached, and then determining the cluster center according to the final centroid of each cluster; determining extreme climate scenarios and typical climate scenarios according to the distribution of historical climate data in the cluster, including: calculating the Euclidean distance from each historical climate data to the cluster center of the cluster to which it belongs; for each cluster, screening a preset number of historical climate data with the largest Euclidean distance as special climate data.

[0112] Figure 2 A schematic diagram of a process for determining extreme climate scenarios and typical climate scenarios based on historical climate data is shown in an embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0113] Step S201, input historical climate data, and give a range of cluster numbers; the initial cluster number can be 1, and the maximum cluster number can be k max ;

[0114] Step S202, input the number of clusters;

[0115] Step S203, call the k-means algorithm to perform clustering, and calculate the sum of squared errors of the Euclidean distance from the historical climate data in each cluster to the centroid of the cluster, and then calculate the SSE of all historical climate data. k , add 1 to the number of clusters and calculate SSE according to the above steps k+1 ;

[0116] Step S204, for k=1, 2, ..., k max -1Calculate K = SSE k+1 / SSE k , and determine that the k+1 corresponding to the smallest K is the target number of clusters for the optimal clustering.

[0117] Step S205, randomly selecting a historical climate data from the historical climate data as the first initial centroid c1;

[0118] Step S206, calculating the Euclidean distance from other historical climate data to c1, and then determining the probability of other historical climate data being selected as the second initial centroid; the greater the Euclidean distance from other historical climate data to c1, the higher the probability of being selected as the second initial centroid;

[0119] Step S207, determining a second initial centroid c2;

[0120] Step S208, determining the Euclidean distance from other historical climate data to the nearest initial centroid, and then determining the probability of other historical climate data being selected as the next initial centroid; the greater the Euclidean distance from other historical climate data to the nearest initial centroid, the higher the probability of being selected as the next initial centroid;

[0121] Step S209, determining the next initial centroid;

[0122] Step S210, determine whether the number of initial centroids reaches the target number of clusters; if not, go to step S208, if yes, go to step S211;

[0123] Step S211, calculating the Euclidean distance from each historical climate data to k+1 initial centroids, and classifying each historical climate data to the initial centroid with the smallest Euclidean distance;

[0124] Step S212, for each cluster, recalculate the centroid using the historical climate data classified into the cluster, and iterate until a stop condition is reached;

[0125] Step S213, determining the cluster center according to the final centroid of each cluster;

[0126] Step S214, calculating the Euclidean distance between each historical climate data and the cluster center of the cluster to which it belongs;

[0127] Step S215, for each cluster, selecting a preset number of historical climate data with the largest Euclidean distance as special climate data; the preset number may be 1, 2, etc.;

[0128] Step S216, determining an extreme climate scene according to the date corresponding to the special climate data;

[0129] Step S217, determining a typical climate scene according to the extreme climate scene.

[0130] The following Table 1 exemplifies the dates corresponding to extreme climate scenarios determined based on the distribution of historical climate data in clusters using the elbow method and the k-means++ algorithm to cluster historical climate data in 2022 or 2023.

[0131] Table 1:

[0132] 2022 2023 date March 22 January 13 date April 13 January 25 date April 14 June 24 date September 15 July 16

[0133] In some embodiments of the present application, in the above method, step S130, constructing a load condition probability distribution model based on multidimensional wind and solar load historical data, including: using D-vine structure and non-parametric kernel density estimation method to perform correlation modeling on the multidimensional wind and solar load historical data to obtain a load condition probability distribution model.

[0134] Specifically, the D-vine structure and non-parametric kernel density estimation method are used to model the correlation of multidimensional wind and solar load historical data, and a load condition probability distribution model is obtained, including: using the non-parametric kernel density estimation method to perform probability integral transformation on the wind and solar load historical data; based on the sample set after the probability integral transformation, the function type and parameters of each Pair Copula in the D-vine structure are optimized; based on the optimized D-vine structure, a load condition probability distribution model with the output of the wind turbine and the output of the photovoltaic generator as conditional variables is established.

[0135] Read in multi-dimensional wind and solar load historical data X 1i ,X 2i ,…,X di , d≥2, i=1,2,...,N.

[0136] Calculate the marginal probability density function f using nonparametric kernel density estimation k (x k ) and the marginal cumulative distribution function F k (x k ), where x k represents a random variable, k=1,2,…,d.

[0137]

[0138] Where h represents bandwidth, X ki represents the random variable x k The i-th historical sample of , K represents the kernel function.

[0139] Let u k =F k (x k), after the multi-dimensional wind and solar load historical data is introduced, the uniform distribution sample set U1=[U1,U2,…,U d ].

[0140] The D-vine structure expression is:

[0141]

[0142] Where c represents the Copula density function of the D-vine structure,

[0143] The Copula density function of the L1 layer of the D-vine structure was fitted according to the sample set U1, and the parameters were obtained by maximum likelihood estimation. The fitting degree of the parameters to the sample set U1 was evaluated by the Euclidean distance method, and the optimal Copula density function of the L1 layer was determined.

[0144] The sample set U2 is obtained according to the sample set U1 and the sample set of the second layer conditional distribution generated according to the D-vine structure expression. The L2-layer Copula density function of the D-vine structure is fitted according to the sample set U2, and the parameters are obtained by maximum likelihood estimation. The fitting degree of the parameters to the sample set U2 is evaluated according to the Euclidean distance method, and the optimal Copula density function of the L2 layer is determined.

[0145] Based on the same method as the above steps, the sample set U(d-2) of the previous layer and the sample set of the conditional distribution of this layer generated according to the D-vine structure expression are used to obtain the sample set U(d-1). d-1 The layer Copula density function is fitted, and the parameters are obtained by maximum likelihood estimation. The fitting degree of the parameters to the sample set U(d-1) is evaluated according to the Euclidean distance method to determine the Lth d-1 The optimal Copula density function for each layer.

[0146] Collect data and generate data X1, X2, …, X containing multi-dimensional wind-light-load correlation d .

[0147] The sampling function is:

[0148]

[0149] Among them, W1, W2, …, W d is a random number generated by the Monte Carlo method and follows a uniform distribution on [0,1]. -1 Represents the inverse of each marginal cumulative distribution function.

[0150] Through the above steps, a load condition probability distribution model is established with the output of the wind turbine generator set and the output of the photovoltaic generator set as conditional variables.

[0151] Figure 3 A schematic diagram showing the output of a wind turbine generator set and a photovoltaic generator set on June 24, 2023 proposed in an embodiment of the present application is shown.

[0152] In some embodiments of the present application, in the above method, step S140, according to the multi-dimensional wind and solar load historical data, the basic data of the power grid, and the load condition probability distribution model, calculates the wind and solar capacity credibility under extreme climate scenarios and the wind and solar capacity credibility under typical climate scenarios, including: taking the power shortage expectation value as the reliability index; determining the benchmark reliability index of the system at each moment that does not include the wind turbine generator set and the photovoltaic power generation group according to the multi-dimensional wind and solar load historical data and the basic data of the power grid; determining the system that includes the wind turbine generator set and the photovoltaic power generation group according to the multi-dimensional wind and solar load historical data, the basic data of the power grid, and the load condition probability distribution model. The dynamic reliability index of the system of the unit at each moment; when the difference between the dynamic reliability index at the corresponding moment and the benchmark reliability index is less than the preset threshold, the adjustment amount of the startup capacity of the conventional unit is determined as the wind and solar operation credible capacity at each moment; according to the date corresponding to the extreme climate scenario and the date corresponding to the typical climate scenario, based on the wind and solar operation credible capacity at each moment, the wind and solar operation credible capacity under the extreme climate scenario and the wind and solar operation credible capacity under the typical climate scenario are calculated respectively; according to the wind and solar operation credible capacity and the wind and solar installed capacity, the wind and solar capacity credibility under the extreme climate scenario and the wind and solar capacity credibility under the typical climate scenario are calculated.

[0153] In this embodiment, the power shortage expected value is used as the reliability indicator.

[0154] Specifically, based on multi-dimensional wind and solar load historical data and basic power grid data, the benchmark reliability indicators of the system excluding wind turbines and photovoltaic generators at each moment are determined, including: determining the startup capacity of conventional units based on the installed capacity of conventional units, forced outage rate, peak load, and rotating reserve rate; and using the Monte Carlo method to determine the benchmark reliability indicators at each moment based on the startup capacity of conventional units and the load data at each moment.

[0155] First, according to the installed capacity of conventional units, forced outage rate, peak load and spinning reserve rate, the startup capacity of conventional units is determined. The startup capacity of conventional units is the startup capacity P of the power system when wind turbines and photovoltaic generators are removed. old .

[0156] According to the conventional unit startup capacity P oldAs well as the load data at each moment, the Monte Carlo method is used to calculate EENS1 without the wind and solar power output.

[0157] Specifically, the dynamic reliability index of the system including wind turbines and photovoltaic generators is determined based on multi-dimensional wind and solar load historical data, basic power grid data, and a load condition probability distribution model, including: adjusting the startup capacity of conventional units by using the dichotomy method or the truncation method; and determining the dynamic reliability index at each moment by using the Monte Carlo method based on the adjusted startup capacity of conventional units, the load condition probability distribution model, and the load data at each moment.

[0158] When the power system is connected to the wind and solar power output, the conventional unit startup capacity is adjusted by the dichotomy method or the truncation method. The adjusted conventional unit startup capacity is P new .

[0159] According to the adjusted conventional unit startup capacity P new As well as the load data at each moment, the Monte Carlo method is used to calculate the EENS2 connected to the wind and solar power output.

[0160] If the difference between EENS1 and EENS2 at the corresponding moment is less than the preset threshold, that is, the difference between the two is within a certain error range, then ΔP = P old -P new As the scenery of the moment operates trusted capacity.

[0161] According to the dates corresponding to extreme climate scenarios and the dates corresponding to typical climate scenarios, based on the wind and solar credible capacity at each moment, the wind and solar credible capacity under extreme climate scenarios and the wind and solar credible capacity under typical climate scenarios can be calculated respectively. Then, based on the wind and solar credible capacity and the wind and solar installed capacity, the wind and solar capacity credibility under extreme climate scenarios and the wind and solar capacity credibility under typical climate scenarios can be calculated.

[0162] Figure 4 A schematic diagram of a process for calculating the credibility of wind and solar capacity in extreme climate scenarios and the credibility of wind and solar capacity in typical climate scenarios according to multi-dimensional wind and solar load history data, basic grid data, and load condition probability distribution model proposed in an embodiment of the present application is shown. Figure 4 As shown, the process includes the following steps:

[0163] Step S401, read in multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model;

[0164] Step S402: Determine the startup capacity P of the conventional units according to the installed capacity of the conventional units, the forced outage rate, the peak load, and the spinning reserve rate. old ;

[0165] Step S403, set the range of sampling times; the maximum sampling times can be N max ;

[0166] Step S404, sampling the nth conventional unit status and the nth load data;

[0167] Step S405, determine whether n is equal to N max ; n = N max ? If yes, go to step S406, if not, add 1 to n and go to step S404;

[0168] Step S406, calculating EENS1;

[0169] Step S407, using the binary method to adjust the startup capacity of the conventional units, the adjusted startup capacity of the conventional units is P new ;

[0170] Step S408, sampling of the state of the conventional unit after the nth adjustment, sampling of the load data for the nth time, and sampling of the wind and solar output based on the load condition probability distribution model for the nth time;

[0171] Step S409, determine whether n is equal to N max ; n = N max ? If yes, go to step S410, if not, n is increased by 1 and go to step S408;

[0172] Step S410, calculate EENS2

[0173] Step S411, determine whether the difference between EENS2 and EENS1 is less than a preset threshold; |EENS2-EENS1|<α? If yes, go to step S412, if not, go to step S407;

[0174] Step S412, calculate ΔP=P old -P new , ΔP is taken as the wind and solar operation reliable capacity at that moment;

[0175] Step S413, according to the date corresponding to the extreme climate scenario, based on the wind and solar operation reliable capacity at each moment, calculate the wind and solar operation reliable capacity under the extreme climate scenario, and then calculate the wind and solar capacity reliability under the extreme climate scenario according to the wind and solar operation reliable capacity and the wind and solar installed capacity;

[0176] Step S414, according to the date corresponding to the typical climate scene, based on the wind and solar operation reliable capacity at each moment, calculate the wind and solar operation reliable capacity under the typical climate scene, and then calculate the wind and solar capacity credibility under the typical climate scene based on the wind and solar operation reliable capacity and the wind and solar installed capacity.

[0177] Tables 2 to 5 exemplarily show the wind and solar capacity credibility under extreme climate scenarios in 2022, the wind and solar capacity credibility under extreme climate scenarios in 2023, the wind and solar capacity credibility under typical climate scenarios in 2022, and the wind and solar capacity credibility under typical climate scenarios in 2023. Among them, the typical climate scenarios are divided into spring, summer, autumn and winter according to the seasons.

[0178] In the embodiments illustrated in Tables 2 to 5, the installed capacity of conventional units is 26,865 MW and the rotating reserve rate is 5%; the installed capacity of offshore wind turbines is 669 MW, the installed capacity of onshore wind turbines is 401 MW, the installed capacity of distributed photovoltaic generators is 2,696 MW, and the installed capacity of centralized photovoltaic generators is 398 MW; the maximum load is 36,500 MW.

[0179] Table 2 illustrates the credibility of wind and solar capacity under extreme climate scenarios in 2022.

[0180] Table 2:

[0181] March 22 April 13 April 14 September 15 Offshore wind turbines 29.3 83.4 73.0 56.4 Onshore wind turbines 63.1 37.7 36.1 59.3 Distributed photovoltaic generators 13.1 10.6 16.0 25.0 Centralized photovoltaic generator 28.9 16.5 27.9 40.0 Wind turbines 42.0 65.7 57.8 57.0 Photovoltaic generator set 15.2 11.4 17.4 27.1 Wind and solar joint 25.3 30.0 36.1 38.5

[0182] Table 3 shows the credibility of wind and solar capacity under the extreme climate scenario in 2023.

[0183] Table 3:

[0184] January 13 January 25 June 24 July 16 Offshore wind turbines 2.9 0.3 103.8 52.1 Onshore wind turbines 1.0 1.0 75.4 17.9 Distributed photovoltaic generators 9.2 58.9 11.0 37.4 Centralized photovoltaic generator 8.4 86.0 10.9 33.6 Wind turbines 2.2 0.3 93.1 39.7 Photovoltaic generator set 9.1 61.7 11.0 36.7 Wind and solar joint 6.7 40.7 36.8 37.7

[0185] Table 4 illustrates the credibility of wind and solar capacity under typical climate scenarios in 2022.

[0186] Table 4:

[0187] spring summer autumn winter Offshore wind turbines 23.4 16.6 35.5 31.7 Onshore wind turbines 38.3 11.8 25.9 36.6 Distributed photovoltaic generators 47.1 61.2 46.3 36.1 Centralized photovoltaic generator 38.6 46.1 30.5 27.0 Wind turbines 28.3 14.9 30.4 32.6 Photovoltaic generator set 39.1 47.9 32.7 27.2 Wind and solar joint 35.9 35.3 34.6 29.6

[0188] Table 5 illustrates the credibility of wind and solar capacity under typical climate scenarios in 2023.

[0189] Table 5:

[0190] spring summer autumn winter Offshore wind turbines 28.8 26.5 19.8 36.7 Onshore wind turbines 18.1 19.8 16.5 31.8 Distributed photovoltaic generators 26.3 45.8 39.8 24.7 Centralized photovoltaic generator 24.3 56.1 48.7 36.2 Wind turbines 24.1 24.0 17.7 34.7 Photovoltaic generator set 26.0 46.7 40.8 26.1 Wind and solar joint 25.6 39.4 34.0 28.7

[0191] According to the wind and solar capacity credibility results shown in Tables 2 to 5, it can be seen that the range of wind and solar capacity credibility under extreme climate scenarios is significantly larger than that under typical climate scenarios. The wind power capacity credibility under some extreme climate scenarios is more than 90%, and the wind power capacity credibility under some extreme climate scenarios is less than 5%.

[0192] Due to the intermittent, volatile and random nature of wind and solar energy, the reliability of wind and solar energy capacity in extreme climate scenarios and typical climate scenarios is quite different. Therefore, when arranging the power generation plan of the power system in the future, the impact of climate conditions on the reliability of wind and solar energy capacity should be comprehensively considered, and combined with the output forecast of wind turbines and photovoltaic generators, the power generation plan of conventional units can be arranged more reasonably.

[0193] Figure 5 A schematic diagram of a calculation device for the reliability of wind and solar capacity in an extreme climate scenario according to an embodiment of the present application is shown. Figure 5 As shown, the device 500 includes:

[0194] Data unit 510, used to obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data;

[0195] A scene extraction unit 520, for determining extreme climate scenes and typical climate scenes according to historical climate data;

[0196] A model building unit 530 is used to build a load condition probability distribution model based on multi-dimensional wind and solar load historical data;

[0197] The solving unit 540 is used to calculate the credibility of wind and solar capacity under extreme climate scenarios and the credibility of wind and solar capacity under typical climate scenarios based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model.

[0198] In some embodiments of the present application, in the above-mentioned device 500, the scene extraction unit 520 includes: a clustering module, which is used to cluster historical climate data using the elbow method and the k-means++ algorithm; an extraction module, which is used to determine extreme climate scenes and typical climate scenes according to the distribution of historical climate data in the cluster.

[0199] In some embodiments of the present application, in the above-mentioned device 500, the clustering module is also used to: determine the target number of clusters for clustering using the elbow method; determine the initial centroid of each cluster using the k-means++ algorithm; calculate the Euclidean distance from each historical climate data to each initial centroid, classify each historical climate data into the initial centroid with the smallest Euclidean distance, recalculate the centroid for each cluster using the historical climate data classified into the cluster, and determine the cluster center according to the final centroid of each cluster after iterating until the stopping condition is reached.

[0200] In some embodiments of the present application, in the above-mentioned device 500, the extraction module is also used to: calculate the Euclidean distance of each historical climate data to the cluster center of the cluster to which it belongs; for each cluster, screen a preset number of historical climate data with the largest Euclidean distance as special climate data; determine the extreme climate scene according to the date corresponding to the special climate data, and determine the typical climate scene according to the extreme climate scene.

[0201] In some embodiments of the present application, in the above-mentioned device 500, the model building unit 530 is specifically used to: use the D-vine structure and non-parametric kernel density estimation method to perform correlation modeling on multi-dimensional wind and solar load historical data to obtain a load condition probability distribution model.

[0202] In some embodiments of the present application, in the above-mentioned device 500, the model building unit 530 is also specifically used to: use a non-parametric kernel density estimation method to perform a probability integral transformation on the historical data of wind and solar loads; based on the sample set after the probability integral transformation, optimize the function type and parameters of each Pair Copula in the D-vine structure; based on the optimized D-vine structure, establish a load conditional probability distribution model with the output of the wind turbine and the output of the photovoltaic turbine as conditional variables.

[0203] In some embodiments of the present application, in the above-mentioned device 500, the solution unit 540 includes: a baseline reliability index determination module, which is used to determine the baseline reliability index of the system at each moment that does not include the wind turbine generator set and the photovoltaic generator set according to the multi-dimensional wind-solar load historical data and the basic data of the power grid, wherein the power shortage expectation value is used as the reliability index; a dynamic reliability index determination module, which is used to determine the dynamic reliability index of the system at each moment that includes the wind turbine generator set and the photovoltaic generator set according to the multi-dimensional wind-solar load historical data, the basic data of the power grid, and the load condition probability distribution model, wherein the power shortage expectation value is used as the reliability index; The operation credible capacity calculation module is used to determine the adjustment amount of the conventional unit startup capacity as the wind and solar operation credible capacity at each moment when the gap between the dynamic reliability index at the corresponding moment and the benchmark reliability index is less than a preset threshold value. According to the date corresponding to the extreme climate scenario and the date corresponding to the typical climate scenario, based on the wind and solar operation credible capacity at each moment, the wind and solar operation credible capacity under the extreme climate scenario and the wind and solar operation credible capacity under the typical climate scenario are calculated respectively; the wind and solar capacity credibility calculation module is used to calculate the wind and solar capacity credibility under the extreme climate scenario and the wind and solar capacity credibility under the typical climate scenario according to the wind and solar operation credible capacity and the wind and solar installed capacity.

[0204] In some embodiments of the present application, in the above-mentioned device 500, the benchmark reliability index determination module is also used to: determine the conventional unit startup capacity based on the conventional unit installed capacity, forced outage rate, peak load, and rotating reserve rate; and determine the benchmark reliability index at each moment using the Monte Carlo method based on the conventional unit startup capacity and the load data at each moment.

[0205] In some embodiments of the present application, in the above-mentioned device 500, the dynamic reliability index determination module is also used to: adjust the startup capacity of the conventional unit by using the bisection method or the truncation method; and determine the dynamic reliability index at each moment by using the Monte Carlo method based on the adjusted startup capacity of the conventional unit, the load condition probability distribution model, and the load data at each moment.

[0206] It should be noted that the above-mentioned calculation device 500 for the credibility of wind and solar capacity considering extreme climate scenarios can implement the above-mentioned calculation method for the credibility of wind and solar capacity considering extreme climate scenarios one by one, which will not be repeated here.

[0207] Figure 6 A schematic diagram showing the structure of a computer device according to an embodiment of the present application is shown. Figure 6 As shown, the internal structure of the computer device may include a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, the functions or steps of the calculation method of the credibility of wind and solar capacity considering extreme climate scenarios are implemented.

[0208] In one embodiment, the computer device provided by the present application includes a memory and a processor, the memory stores a database and a computer program that can be run on the processor, and the processor implements the following steps when executing the computer program:

[0209] Obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data;

[0210] Determine extreme climate scenarios and typical climate scenarios based on historical climate data;

[0211] Construct a load condition probability distribution model based on multi-dimensional wind and solar load historical data;

[0212] Based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model, the credibility of wind and solar capacity under extreme climate scenarios and the credibility of wind and solar capacity under typical climate scenarios are calculated.

[0213] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0214] Obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data;

[0215] Determine extreme climate scenarios and typical climate scenarios based on historical climate data;

[0216] Construct a load condition probability distribution model based on multi-dimensional wind and solar load historical data;

[0217] Based on multi-dimensional wind and solar load historical data, basic grid data, and load condition probability distribution model, the credibility of wind and solar capacity under extreme climate scenarios and the credibility of wind and solar capacity under typical climate scenarios are calculated.

[0218] It should be noted that the above functions or steps that can be implemented by the computer device or computer-readable storage medium can refer to the relevant description in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0219] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0220] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0221] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for calculating the credibility of wind and solar capacity under extreme climate scenarios, characterized in that: The method comprises: Obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data; determining extreme climate scenarios and typical climate scenarios based on the historical climate data; Constructing a load condition probability distribution model based on the multi-dimensional wind and solar load historical data; According to the multi-dimensional wind and solar load historical data, the basic data of the power grid, and the load condition probability distribution model, the wind and solar capacity credibility under the extreme climate scenario and the wind and solar capacity credibility under the typical climate scenario are calculated.

2. The method for calculating the reliability of wind and solar capacity under extreme climate scenarios according to claim 1 is characterized in that: Determining extreme climate scenarios and typical climate scenarios based on the historical climate data includes: The historical climate data are clustered using the elbow method and the k-means++ algorithm; The extreme climate scenario and the typical climate scenario are determined according to the distribution of the historical climate data in the cluster.

3. The method for calculating the reliability of wind and solar capacity under extreme climate scenarios according to claim 2 is characterized in that: The clustering of the historical climate data using the elbow method and the k-means++ algorithm includes: Using the elbow method to determine the target number of clusters for the clustering; Using the k-means++ algorithm to determine the initial centroid of each cluster; Calculating the Euclidean distance from each of the historical climate data to each of the initial centroids, classifying each of the historical climate data to the initial centroid with the smallest Euclidean distance, recalculating the centroid for each cluster using the historical climate data classified in the cluster, iterating until a stop condition is reached, and then determining the cluster center according to the final centroid of each cluster; The determining the extreme climate scenario and the typical climate scenario according to the distribution of the historical climate data in the cluster includes: Calculating the Euclidean distance between each of the historical climate data and the cluster center of the cluster to which it belongs; For each of the clusters, selecting a preset number of the historical climate data with the largest Euclidean distance as special climate data; The extreme climate scene is determined according to the date corresponding to the special climate data, and the typical climate scene is determined according to the extreme climate scene.

4. The method for calculating the reliability of wind and solar capacity under extreme climate scenarios according to claim 1 is characterized in that: The constructing of a load condition probability distribution model according to the multi-dimensional wind and solar load historical data includes: The D-vine structure and non-parametric kernel density estimation method are used to perform correlation modeling on the multi-dimensional wind-solar load historical data to obtain the load condition probability distribution model.

5. The method for calculating the reliability of wind and solar capacity under extreme climate scenarios according to claim 4 is characterized in that: The D-vine structure and non-parametric kernel density estimation method are used to perform correlation modeling on the multi-dimensional wind-solar load historical data to obtain the load condition probability distribution model, including: Using the non-parametric kernel density estimation method to perform probability integral transformation on the wind and solar load historical data; Based on the sample set after the probability integral transformation, optimizing the function type and parameters of each pair copula in the D-vine structure; Based on the optimized D-vine structure, the load condition probability distribution model is established with the output of the wind turbine generator set and the output of the photovoltaic generator set as conditional variables.

6. The method for calculating the reliability of wind and solar capacity under extreme climate scenarios according to claim 1 is characterized in that: The calculating the wind / solar capacity credibility under the extreme climate scenario and the wind / solar capacity credibility under the typical climate scenario according to the multi-dimensional wind / solar load history data, the basic grid data, and the load condition probability distribution model includes: The expected value of power shortage is used as a reliability indicator; Determine the benchmark reliability index of the system at each moment that does not include the wind turbine generator set and the photovoltaic generator set according to the multi-dimensional wind, solar and load historical data and the basic data of the power grid; Determine the dynamic reliability index of the system including the wind turbine generator set and the photovoltaic generator set at each moment according to the multi-dimensional wind-solar-load historical data, the basic grid data, and the load condition probability distribution model; When the difference between the dynamic reliability index at the corresponding moment and the benchmark reliability index is less than a preset threshold, the adjustment amount of the startup capacity of the conventional unit is determined as the wind and solar operation reliable capacity at each moment; According to the date corresponding to the extreme climate scenario and the date corresponding to the typical climate scenario, based on the wind and solar operation reliable capacity at each moment, respectively calculate the wind and solar operation reliable capacity under the extreme climate scenario and the wind and solar operation reliable capacity under the typical climate scenario; According to the wind-solar operation reliable capacity and the wind-solar installed capacity, the wind-solar capacity credibility under the extreme climate scenario and the wind-solar capacity credibility under the typical climate scenario are calculated.

7. The method for calculating the reliability of wind and solar capacity under extreme climate scenarios according to claim 6 is characterized in that: Determining the benchmark reliability index of the system excluding the wind turbine generator set and the photovoltaic generator set at each moment according to the multi-dimensional wind, solar and load historical data and the basic grid data includes: Determine the operating capacity of conventional units based on the installed capacity of conventional units, forced outage rate, peak load, and spinning reserve rate; Determine the benchmark reliability index at each moment using the Monte Carlo method according to the startup capacity of the conventional units and the load data at each moment; Determining the dynamic reliability index of the system including the wind turbine generator set and the photovoltaic generator set at each moment according to the multi-dimensional wind-solar-load historical data, the basic grid data, and the load condition probability distribution model includes: The start-up capacity of the conventional unit is adjusted by using the dichotomy method or the truncation method; The dynamic reliability index at each moment is determined by using the Monte Carlo method according to the adjusted startup capacity of the conventional units, the load condition probability distribution model, and the load data at each moment.

8. A computing device for the reliability of wind and solar capacity considering extreme climate scenarios, characterized in that: The device comprises: Data unit, used to obtain historical climate data, multi-dimensional wind, solar and load historical data, and basic power grid data; A scene extraction unit, used to determine extreme climate scenes and typical climate scenes according to the historical climate data; A model building unit, used to build a load condition probability distribution model according to the multi-dimensional wind and solar load historical data; A solution unit is used to calculate the credibility of the wind and solar capacity under the extreme climate scenario and the credibility of the wind and solar capacity under the typical climate scenario based on the multi-dimensional wind and solar load historical data, the basic data of the power grid, and the load condition probability distribution model.

9. The device for calculating the reliability of wind and solar capacity in extreme climate scenarios according to claim 8, characterized in that: The scene extraction unit comprises: A clustering module, for clustering the historical climate data using an elbow method and a k-means++ algorithm; An extraction module is used to determine the extreme climate scene and the typical climate scene according to the distribution of the historical climate data in the cluster.

10. The device for calculating the reliability of wind and solar capacity in extreme climate scenarios according to claim 9, characterized in that: The clustering module is further used for: Using the elbow method to determine the target number of clusters for the clustering; Using the k-means++ algorithm to determine the initial centroid of each cluster; Calculating the Euclidean distance from each of the historical climate data to each of the initial centroids, classifying each of the historical climate data to the initial centroid with the smallest Euclidean distance, recalculating the centroid for each cluster using the historical climate data classified in the cluster, iterating until a stop condition is reached, and then determining the cluster center according to the final centroid of each cluster; The extraction module is further used for: Calculating the Euclidean distance between each of the historical climate data and the cluster center of the cluster to which it belongs; For each of the clusters, selecting a preset number of the historical climate data with the largest Euclidean distance as special climate data; The extreme climate scene is determined according to the date corresponding to the special climate data, and the typical climate scene is determined according to the extreme climate scene.

11. The device for calculating the reliability of wind and solar capacity in extreme climate scenarios according to claim 8, characterized in that: The model building unit is used to: The D-vine structure and non-parametric kernel density estimation method are used to perform correlation modeling on the multi-dimensional wind-solar load historical data to obtain the load condition probability distribution model.

12. The device for calculating the reliability of wind and solar capacity in extreme climate scenarios according to claim 11, characterized in that: The model building unit is further used for: Using the non-parametric kernel density estimation method to perform probability integral transformation on the wind and solar load historical data; Based on the sample set after the probability integral transformation, optimizing the function type and parameters of each pair copula in the D-vine structure; Based on the optimized D-vine structure, the load condition probability distribution model is established with the output of the wind turbine generator set and the output of the photovoltaic generator set as conditional variables.

13. The device for calculating the reliability of wind and solar capacity in extreme climate scenarios according to claim 8, characterized in that: The solving unit comprises: A benchmark reliability index determination module, used to determine the benchmark reliability index of the system excluding the wind turbine generator set and the photovoltaic generator set at each moment according to the multi-dimensional wind, solar and load historical data and the basic data of the power grid, wherein the power shortage expected value is used as the reliability index; A dynamic reliability index determination module, used to determine the dynamic reliability index of the system including the wind turbine generator set and the photovoltaic generator set at each moment according to the multi-dimensional wind-solar-load historical data, the basic data of the power grid, and the load condition probability distribution model; A wind-solar operation credible capacity calculation module is used to determine the adjustment amount of the conventional unit startup capacity as the wind-solar operation credible capacity at each moment when the difference between the dynamic reliability index at the corresponding moment and the benchmark reliability index is less than a preset threshold value, and to calculate the wind-solar operation credible capacity under the extreme climate scenario and the wind-solar operation credible capacity under the typical climate scenario based on the wind-solar operation credible capacity at each moment according to the date corresponding to the extreme climate scenario and the date corresponding to the typical climate scenario; The wind and solar capacity credibility calculation module is used to calculate the wind and solar capacity credibility under the extreme climate scenario and the wind and solar capacity credibility under the typical climate scenario according to the wind and solar operation credibility capacity and the wind and solar installed capacity.

14. The device for calculating the reliability of wind and solar capacity in extreme climate scenarios according to claim 13, characterized in that: The benchmark reliability index determination module is also used for: Determine the operating capacity of conventional units based on the installed capacity of conventional units, forced outage rate, peak load, and spinning reserve rate; Determine the benchmark reliability index at each moment using the Monte Carlo method according to the startup capacity of the conventional units and the load data at each moment; The dynamic reliability index determination module is further used to: adjust the startup capacity of the conventional unit by using the dichotomy method or the truncation method; The dynamic reliability index at each moment is determined by using the Monte Carlo method according to the adjusted startup capacity of the conventional units, the load condition probability distribution model, and the load data at each moment.

15. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for calculating the credibility of wind and solar capacity considering extreme climate scenarios as described in any one of claims 1 to 7 are implemented.

16. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is instructed by the processor, it implements the steps of the method for calculating the credibility of wind and solar capacity considering extreme climate scenarios as described in any one of claims 1 to 7.

Images