Method and device for generating typical scene of wind and light

Abstract

The invention relates to the technical field of new energy modeling, in particular to a wind and light typical scene generation method and device.The method comprises the steps that firstly, wind and light historical output data of a target wind and light power station are obtained, and a wind and light energy output edge distribution function of the target wind and light power station is calculated based on the wind and light historical output data; based on the wind and light energy output edge distribution function, using a mixed Copula function to construct a wind and light output joint distribution model; carrying out sampling and splicing processing on the joint distribution model of the wind and light output to generate a wind and light typical scene set; and finally, calculating the occurrence probability of each target scene in the wind-light typical scene set, calculating the correlation between the target scene and the wind-light historical output data, and selecting a final wind-light typical scene based on the occurrence probability and the correlation coefficient. The deviation between the generated wind and light typical scene and the actual situation is small, consideration on the correlation of the original wind and light data is combined, and the operation efficiency and safety of the power system are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy modeling, and in particular to a method and device for generating typical wind and solar scenes. Background Art

[0002] At present, the large-scale development and utilization of wind and solar green energy (such as wind power generation and photovoltaic power generation) has become an important part of the national energy strategy. As renewable energy, wind and solar energy (wind power and solar energy) has huge development potential, but its output is affected by natural factors, such as changes in wind speed and direction, and changes in sunshine intensity and duration, so it has strong uncertainty.

[0003] In order to effectively manage and utilize these energies, the power system needs to quantify the uncertainty of wind and solar energy output. This helps the power system to conduct more accurate energy management, risk assessment and decision-making, thereby ensuring the stable operation of the power grid and the efficient use of energy. However, the existing technology usually uses a single type of Copula function to construct typical wind and solar time series scenarios. Copula function is a mathematical tool used to describe the correlation between multidimensional random variables. However, a single type of Copula function is often difficult to fully describe the complex correlation relationship between wind and solar output data. This may cause the generated wind and solar typical scenarios to deviate from the actual situation, thereby affecting the operating efficiency and safety of the power system. In addition, the generation and selection of existing wind and solar typical scenarios lack consideration of the correlation with the original wind and solar data, which is not conducive to the safe and economical operation of the power grid. Summary of the invention

[0004] In view of this, the present invention provides a method and device for generating typical wind and solar power scenes, so as to solve the problem that the typical wind and solar power scenes generated by the existing methods deviate from the actual situation, thereby affecting the operating efficiency and safety of the power system.

[0005] In a first aspect, the present invention provides a method for generating a typical scenery scene, the method comprising:

[0006] Acquire historical wind and solar power output data of a target wind and solar power station, and calculate a wind and solar energy output marginal distribution function of the target wind and solar power station based on the historical wind and solar power output data;

[0007] Based on the wind and solar energy output marginal distribution function, a hybrid Copula function is used to construct a joint distribution model of wind and solar output;

[0008] Sampling and splicing the joint distribution model of wind and solar power output to generate a set of typical wind and solar sceneries;

[0009] The occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind and solar power output data are calculated, and a final typical scene is selected based on the occurrence probability and the correlation coefficient.

[0010] The above scheme is based on the marginal distribution function of wind and solar energy output, and uses a hybrid Copula function to construct a joint distribution model of wind and solar output. It is not limited to a single Copula function, but can adaptively adjust the model parameters according to the historical wind and solar output data. It can flexibly change the parameters in the hybrid Copula function according to the specific situation of the actual wind and solar output data, so that the constructed model can better fit the data characteristics, thereby generating a joint distribution model of wind and solar output with the best fitting effect. Since the wind and solar time series joint output model with the best fitting effect can be generated, the most representative wind and solar typical scene is selected in combination with the probability of the occurrence of typical scenes, so it has higher accuracy in reflecting the actual wind and solar output in each region, and can more accurately capture the different characteristics and changing laws of wind and solar output caused by differences in geographical, climatic and other factors in different regions. In addition, the most representative wind and solar typical scenes selected by the above scheme and the related probability and correlation coefficient data can provide reliable data support for the planning, scheduling, risk assessment and other aspects of the power system.

[0011] In an optional implementation, the calculating the wind-solar energy output marginal distribution function of the target wind-solar power station based on the wind-solar historical output data includes:

[0012] According to the historical wind and solar power output data, a probability density function of each time period in the target period is established respectively; the probability density function includes a wind power probability density function and a photovoltaic output probability density function;

[0013] The probability density function is integrated to obtain the wind-solar energy output marginal distribution function of the target wind-solar power station; the wind-solar energy output marginal distribution function includes a wind power marginal distribution function and a photovoltaic marginal distribution function.

[0014] In an optional implementation, the wind-solar energy output marginal distribution function is based on the wind-solar energy output marginal distribution function, and a mixed Copula function is used to construct a wind-solar energy output joint distribution model, including:

[0015] Obtaining weight parameters and dependent parameters corresponding to each function in the hybrid Copula function to be constructed; the hybrid Copula function includes a Gumbel Copula function, a Clayton Copula function and a Frank Copula function;

[0016] Constructing the hybrid Copula function based on the weight parameters and dependent parameters corresponding to the functions;

[0017] The wind-solar energy output marginal distribution function and the mixed Copula function are combined to construct the wind-solar output joint distribution model.

[0018] In an optional implementation, the sampling and splicing processing of the joint distribution model of the wind and solar output to generate a set of typical wind and solar sceneries includes:

[0019] Based on the wind-solar power joint distribution model, a wind-solar power joint probability distribution function is obtained;

[0020] Sampling each night time period based on the wind power output marginal distribution function to obtain wind power sampling points in each night time period;

[0021] For the wind power sampling points in the night time period, inverse transformation is performed using the wind-solar joint probability distribution function to obtain typical output values ​​of night wind power in each night time period;

[0022] Based on the hybrid Copula function, the wind-solar joint probability distribution function of each daytime period is sampled to obtain wind power and photovoltaic sampling points of each daytime period;

[0023] For the wind power and photovoltaic sampling points during the daytime period, the typical output value of daytime wind power and the typical output value of daytime photovoltaic are generated by inverting the wind-solar joint probability distribution function.

[0024] The typical output values ​​of nighttime wind power, the typical output values ​​of daytime wind power and the typical output values ​​of daytime photovoltaic power in each period of daytime and nighttime are spliced ​​to generate a typical scene set of wind and solar power for the whole day.

[0025] In an optional implementation, the calculating the occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind power output data, and selecting the final typical scene based on the occurrence probability and the correlation coefficient, includes:

[0026] Obtaining an optimal number of clusters, and clustering the typical wind-solar scene set based on the optimal number of clusters to obtain each cluster, as well as the number of samples and cluster centers of each cluster; the cluster centers indicate representative wind-solar output scenes corresponding to each cluster;

[0027] For each of the clusters, based on the number of samples of the cluster and the total number of samples of the typical scene set of wind and solar power, the occurrence probability of the representative wind and solar power output scene corresponding to each cluster is obtained;

[0028] Calculating the correlation coefficient between the representative wind / solar power output scene corresponding to each cluster and the wind / solar power historical output data;

[0029] In combination with the occurrence probability and the correlation coefficient, the representativeness of the representative wind-solar output scenes corresponding to the respective clusters is obtained, and the representative wind-solar output scene with the highest representativeness is determined as the final wind-solar typical scene.

[0030] In an optional implementation manner, obtaining the optimal number of clusters includes:

[0031] Acquire a value range of the number of clusters, perform clustering processing on the typical scene set based on the value range of the number of clusters, and calculate the sum of squared errors corresponding to each number of clusters in the value range;

[0032] A relationship curve between each cluster number in the value range and the error sum corresponding to the cluster number is constructed, and a target downward trend turning point of the error sum in the relationship curve is determined as the optimal cluster number correlation coefficient.

[0033] In a second aspect, the present invention provides a typical scene generation device, the device comprising:

[0034] An edge distribution function acquisition module is used to acquire the wind and solar historical output data of the target wind and solar power station, and calculate the wind and solar energy output edge distribution function of the target wind and solar power station based on the wind and solar historical output data;

[0035] A joint distribution model building module, used to build a joint distribution model of wind and solar power output using a mixed Copula function based on the wind and solar power output marginal distribution function;

[0036] A wind-solar typical scene set generation module, used for sampling and splicing the wind-solar joint distribution model to generate a wind-solar typical scene set;

[0037] The final typical scene generation module is used to calculate the occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind power output data, and select the final typical scene based on the occurrence probability and the correlation coefficient.

[0038] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes a method for generating a typical scenery scene according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute a method for generating a typical scenery scene according to the first aspect or any corresponding embodiment thereof.

[0040] In a fifth aspect, the present invention provides a computer program product, including computer instructions, which are used to enable a computer to execute a method for generating a typical scenery scene according to the first aspect or any corresponding embodiment thereof.

[0041] The technical solution provided by the present invention may include the following beneficial effects:

[0042] The present invention is based on the marginal distribution function of wind and solar energy output, and uses a hybrid Copula function to construct a joint distribution model of wind and solar output. It is not limited to a single Copula function, but can adaptively adjust the model parameters according to the historical wind and solar output data. It can flexibly change the various parameters in the hybrid Copula function according to the specific situation of the actual wind and solar output data, so that the constructed model can better fit the data characteristics, thereby generating a joint distribution model of wind and solar output with the best fitting effect. Since the present invention can generate a wind and solar time series joint output model with the best fitting effect, and select the most representative wind and solar typical scene in combination with the probability of occurrence of typical scenes, it has higher accuracy in reflecting the actual wind and solar output of each region, and can more accurately capture the different characteristics and changing laws of wind and solar output caused by differences in geographical, climatic and other factors in different regions. In addition, the most representative wind and solar typical scenes selected by the present invention and the related probability and correlation coefficient data can provide reliable data support for the planning, scheduling, risk assessment and other aspects of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 is a flow chart of a method for generating a typical scenery scene according to an embodiment of the present invention;

[0045] Figure 2 is a flow chart of another method for generating a typical scenery scene according to an embodiment of the present invention;

[0046] Figure 3 is a flow chart of another typical scenery scene generation method according to an embodiment of the present invention;

[0047] Figure 4 is a schematic diagram of typical outputs at different time periods during the day and at night according to an embodiment of the present invention;

[0048] Figure 5 is a schematic diagram of a probability curve of photovoltaic output edge distribution of a new energy base according to an embodiment of the present invention;

[0049] Figure 6 is a schematic diagram of a probability curve of wind power output edge distribution of a new energy base according to an embodiment of the present invention;

[0050] Figure 7 is a schematic diagram of typical scene results obtained after K-means clustering processing according to an embodiment of the present invention;

[0051] Figure 8 is a structural block diagram of a typical scenery scene generation device according to an embodiment of the present invention;

[0052] Fig. 9 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0054] According to an embodiment of the present invention, an embodiment of a method for generating a typical scenery scene is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0055] In this embodiment, a method for generating a typical scenery scene is provided. Figure 1 is a flow chart of a typical scene generation method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0056] Step S101, obtaining historical wind and solar power output data of a target wind and solar power station, and calculating a wind and solar energy output marginal distribution function of the target wind and solar power station based on the historical wind and solar power output data.

[0057] Furthermore, this embodiment obtains historical wind power output data and photovoltaic output data from the target wind and solar power stations (i.e., wind power stations and photovoltaic power stations). These data usually include wind power output values ​​and photovoltaic output values ​​for each period of 24 hours in a day, covering a period of time (such as one year) to ensure the representativeness and integrity of the data. Based on the collected historical wind and solar output data, this embodiment establishes the probability density function of wind power output and the probability density function of photovoltaic output, respectively. The probability density function describes the probability that the output of a wind and solar power station falls within a certain range within a given time period. Then, by integrating these probability density functions, the marginal distribution function of wind and solar energy output is obtained. The marginal distribution function of wind and solar energy output indicates the distribution of the output of the target wind and solar power station, including the wind power marginal distribution function and the photovoltaic marginal distribution function.

[0058] Step S102: Based on the wind and solar energy output marginal distribution function, a hybrid Copula function is used to construct a joint distribution model of wind and solar output.

[0059] Furthermore, the hybrid Copula function of this embodiment is composed of multiple Copula functions, each of which corresponds to different weight parameters and dependent parameters, which determine the contribution of each Copula function in the joint distribution model and the correlation between them. This embodiment uses the above parameters to construct a hybrid Copula function, which can better describe the complex correlation relationship between wind and solar output data. The marginal distribution functions of wind power and photovoltaics are combined with the hybrid Copula function to construct a joint distribution model of wind and solar output. The joint distribution model of wind and solar output can describe the joint probability distribution between wind power output and photovoltaic output, that is, the probability of their simultaneous occurrence.

[0060] Step S103, sampling and splicing the joint distribution model of wind and solar power output to generate a set of typical wind and solar scene sets.

[0061] Furthermore, this embodiment samples the joint distribution model of the wind and solar power output to generate a series of wind power and photovoltaic output samples. The sampling can be based on the marginal distribution function of the wind power output for the night time period, and the photovoltaic output at night takes a zero value; for the daytime period, random sampling can be performed based on the mixed Copula model. When splicing, this embodiment splices the wind power output samples and photovoltaic output samples of each period at night and during the day to form a full-day wind and solar power output scene. Through multiple sampling and splicing, multiple typical wind and solar scenes are generated, thereby forming the above-mentioned typical wind and solar scene set.

[0062] Step S104, calculating the occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind power output data, and selecting the final typical scene based on the occurrence probability and the correlation coefficient.

[0063] Furthermore, this embodiment can use the K-means clustering algorithm to cluster the set of typical wind and solar scenes, and determine the optimal number of clusters through the elbow rule. Then calculate the number of samples of the class corresponding to each cluster center as the probability of occurrence of the typical scene. The Pearson correlation coefficient is introduced to calculate the correlation between the original wind and solar historical output data and the typical wind and solar scenes obtained by clustering, that is, the correlation coefficient. The closer the Pearson correlation coefficient is to 1, the stronger the correlation is; the closer it is to -1, the stronger the complementarity is. Combining the probability of occurrence of each target scene and the Pearson correlation coefficient, the most representative final wind and solar typical scene is selected. The final wind and solar typical scene can accurately reflect the wind power and photovoltaic output characteristics of the target wind and solar power station, and provide reliable data support for the optimized operation of the power system.

[0064] In summary, this embodiment is based on the marginal distribution function of wind and solar energy output, and uses a hybrid Copula function to construct a joint distribution model of wind and solar output. It is not limited to a single Copula function, but can adaptively adjust the model parameters according to the historical wind and solar output data. It can flexibly change the parameters in the hybrid Copula function according to the specific situation of the actual wind and solar output data, so that the constructed model can better fit the data characteristics, thereby generating a joint distribution model of wind and solar output with the best fitting effect. Since this embodiment can generate a wind and solar time series joint output model with the best fitting effect, and select the most representative wind and solar typical scene in combination with the probability of occurrence of typical scenes, it has higher accuracy in reflecting the actual wind and solar output in each region, and can more accurately capture the different characteristics and changing laws of wind and solar output caused by differences in geographical, climatic and other factors in different regions. In addition, the most representative wind and solar typical scenes selected by this embodiment and the related probability and correlation coefficient data can provide reliable data support for the planning, scheduling, risk assessment and other aspects of the power system.

[0065] In this embodiment, another method for generating typical scenery scenes is provided. Figure 2 is a flow chart of another typical scene generation method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0066] Step S201, obtaining historical wind and solar power output data of a target wind and solar power station, and calculating a wind and solar energy output marginal distribution function of the target wind and solar power station based on the historical wind and solar power output data.

[0067] In an optional implementation, the step S201 includes:

[0068] According to the historical wind and solar power output data, a probability density function of each time period in the target period is established respectively; the probability density function includes the wind power probability density function and the photovoltaic output probability density function;

[0069] The probability density function is integrated to obtain the wind-solar energy output marginal distribution function of the target wind-solar power station; the wind-solar energy output marginal distribution function includes the wind power marginal distribution function and the photovoltaic marginal distribution function.

[0070] Furthermore, in this embodiment, for the acquired historical wind and solar output data, a probability density function is established for each time period in the target period. The target period can be 24 hours in a day to describe the probability distribution of wind power and photovoltaic output in different time periods of a day. In this embodiment, a day is divided into 24 time periods (each time period corresponds to one hour), and probability density functions of wind power and photovoltaic output are established for each time period, and the probability density function can be implemented by the kernel density estimation method (KDE). The probability density function describes the possibility of wind power and photovoltaic output at different output values, and is the basis for the subsequent calculation of the marginal distribution function. In this embodiment, the established probability density function is integrated, and the marginal distribution function of the wind and solar energy output of the target wind and solar power station can be obtained by integrating the probability density function over its definition domain. The marginal distribution function of wind and solar energy output includes the wind power marginal distribution function and the photovoltaic marginal distribution function. The wind power marginal distribution function describes the cumulative probability of wind power output within the entire possible value range. Similarly, the photovoltaic marginal distribution function describes the cumulative probability of photovoltaic output. These marginal distribution functions provide important basic information for subsequent operations such as constructing a joint distribution model.

[0071] Step S202: Based on the wind and solar energy output marginal distribution function, a hybrid Copula function is used to construct a joint distribution model of wind and solar output.

[0072] In an optional implementation, the step S202 includes:

[0073] Obtain weight parameters and dependent parameters corresponding to each function in the hybrid Copula function to be constructed; the hybrid Copula function includes a Gumbel Copula function, a Clayton Copula function and a Frank Copula function;

[0074] Based on the weight parameters and dependent parameters corresponding to each function, construct the mixed Copula function;

[0075] The wind-solar energy output marginal distribution function and the mixed Copula function are combined to construct the wind-solar output joint distribution model.

[0076] Furthermore, the contribution of each single Copula function in the hybrid Copula function model in this embodiment is determined by a weight parameter. The weight parameter needs to be estimated by a statistical method (such as the least squares method) to ensure that the hybrid Copula function can best fit the actual wind and solar output data. The dependency parameter reflects the strength and pattern of the correlation between wind power and photovoltaic output. Different Copula functions have different dependency parameters, which together determine the shape and characteristics of the hybrid Copula function when describing the correlation between wind and solar output. These dependent parameters also need to be estimated by statistical methods. After obtaining the weight parameters and dependent parameters, this embodiment substitutes these parameters into the expression of the hybrid Copula function to construct a hybrid Copula function. The form of the hybrid Copula function is a weighted sum, in which each single Copula function is weighted according to its weight parameter. Finally, this embodiment combines the constructed hybrid Copula function with the wind and solar energy output marginal distribution function to construct a joint distribution model of wind and solar output. The joint distribution model can simultaneously consider the marginal distribution of wind power and photovoltaic output and the correlation between them, thereby more accurately describing the overall characteristics of wind and photovoltaic output.

[0077] Step S203: sampling and splicing the joint distribution model of wind and solar power output to generate a set of typical wind and solar scene sets.

[0078] In an optional implementation, step S203 includes:

[0079] Based on the joint distribution model of wind and solar power output, a wind and solar power joint probability distribution function is obtained;

[0080] Based on the wind power output marginal distribution function, each night time period is sampled to obtain the wind power sampling point in each night time period;

[0081] For the wind power sampling points in the night time period, the wind-solar joint probability distribution function is used to perform an inverse transformation to obtain the typical output value of night wind power in each night time period;

[0082] Based on the mixed Copula function, the wind-solar joint probability distribution function in each daytime period is sampled to obtain the wind power and photovoltaic sampling points in each daytime period;

[0083] For the wind power and photovoltaic sampling points during the daytime, the typical output values ​​of daytime wind power and daytime photovoltaic power are generated by inverting the wind-solar joint probability distribution function;

[0084] The typical wind power output value at night, the typical wind power output value during the day and the typical photovoltaic output value during the day are spliced ​​together to generate a typical wind and solar scene set for the whole day.

[0085] Further, first, based on the constructed wind-solar joint distribution model, this embodiment can obtain a wind-solar joint probability distribution function. The wind-solar joint probability distribution function describes the joint probability distribution between wind power and photovoltaic output, and is the basis for subsequent sampling and transformation. In the nighttime period, this embodiment samples each nighttime period based on the wind power output marginal distribution function (since the photovoltaic output at night is usually small or zero, the main focus is on wind power), that is, for the nighttime period of a day (such as 20 o'clock in the evening to 8 o'clock in the morning of the next day), sample points are randomly selected from the wind power output marginal distribution function to obtain wind power sampling points in each nighttime period. For the obtained wind power sampling points in the nighttime period, the wind-solar joint probability distribution function is used for inverse transformation. The inverse transformation process is to convert the sampling points in the probability space back to the actual output value space, so as to obtain the typical output value of nighttime wind power in each nighttime period. In the daytime period, this embodiment samples the wind-solar joint probability distribution function of each daytime period (such as 8 o'clock in the morning to 8 o'clock in the evening) based on the mixed Copula function. This means that this embodiment randomly extracts sample points from the joint distribution, and these sample points represent the output values ​​of wind power and photovoltaic power at the same time. Through the daytime sample points obtained by sampling, the typical output values ​​of wind power and photovoltaic power during the day can be obtained by inverting the joint probability distribution function. These values ​​reflect the possible situations of wind power and photovoltaic power output within a given time interval. Finally, this embodiment splices the typical output values ​​of nighttime wind power in each period of the night, the typical output values ​​of wind power in each period of the day, and the typical output values ​​of photovoltaic power, that is, the output values ​​at night and during the day are combined in chronological order to form a 24-hour wind and solar typical scene set. The wind and solar typical scene set contains multiple typical scenes, each of which reflects the possible situations of wind power and photovoltaic power output in different time periods.

[0086] Step S204, obtaining the optimal number of clusters, and clustering the typical wind and solar scene set based on the optimal number of clusters to obtain each cluster, as well as the number of samples and cluster center of each cluster; the cluster center indicates the representative wind and solar power output scene corresponding to each cluster.

[0087] In an optional implementation, the step S204 includes:

[0088] Obtaining a value range of the number of clusters, performing clustering processing on the typical scenery scene set based on the value range of the number of clusters, and calculating the sum of squared errors corresponding to each number of clusters in the value range;

[0089] A relationship curve between each cluster number in the value range and the error sum corresponding to the cluster number is constructed, and a target downward trend turning point of the error sum in the relationship curve is determined as the optimal cluster number.

[0090] Furthermore, the present embodiment first sets a value range for the number of clusters, which can be determined based on experience or actual needs. For example, it can start from a smaller number of clusters (such as 2 or 3) and gradually increase to a larger number to cover all possible reasonable clustering situations. For each number of clusters within the value range, the present embodiment uses the K-means clustering algorithm to cluster the typical scene set of scenery. After each clustering, a corresponding number of clusters, as well as the number of samples and cluster centers of each cluster cluster, are obtained. For each number of clusters, the present embodiment calculates the sum of squared errors (SSE) of the corresponding clustering results. SSE is an important indicator for measuring the clustering effect, which represents the sum of squares of the distances from each point to its cluster center after clustering. The smaller the SSE, the better the clustering effect. The present embodiment plots each number of clusters in the value range and its corresponding SSE into a relationship curve. The relationship curve usually shows a downward trend, that is, as the number of clusters increases, the SSE gradually decreases. In the relationship curve, find the turning point of the downward trend of SSE, and this turning point is the optimal number of clusters. The rate of SSE reduction after the turning point will slow down significantly, and may even fluctuate, which indicates that increasing the number of clusters will have little effect on improving the clustering effect. Afterwards, this embodiment again uses the optimal number of clusters to perform K-means clustering on the set of typical wind and solar scenes to obtain a corresponding number of clusters. For each cluster, calculate the number of samples it contains (that is, the number of typical wind and solar scenes in the cluster). At the same time, calculate the cluster center of each cluster, which represents the average characteristics of all typical wind and solar scenes in the cluster, and can therefore be regarded as a representative wind and solar output scene corresponding to the cluster.

[0091] Step S205 , for each of the clusters, based on the number of samples of the cluster and the total number of samples of the typical wind and solar scene set, obtain the occurrence probability of the representative wind and solar power output scene corresponding to each cluster.

[0092] Furthermore, after completing the clustering processing of the typical wind and solar scene set, this embodiment analyzes the sample data of each cluster cluster to calculate the occurrence probability of the representative wind and solar output scene corresponding to each cluster cluster. This is achieved by counting the number of samples of each cluster cluster and dividing it by the total number of samples of the typical wind and solar scene set.

[0093] Step S206, calculating the correlation coefficient between the representative wind / solar power output scene corresponding to each cluster and the wind / solar power historical output data.

[0094] Furthermore, after obtaining the representative wind and solar power output scene corresponding to each cluster, this embodiment can use statistical methods such as Pearson correlation coefficient to calculate the correlation coefficient between each representative wind and solar power output scene and the original wind and solar power output data. The closer the correlation coefficient is to 1, the stronger the positive correlation between the two is; the closer it is to -1, the stronger the negative correlation between the two is; and the closer it is to 0, the more obvious linear relationship between the two is.

[0095] Step S207 , combining the occurrence probability and the correlation coefficient, obtaining the representativeness of the representative wind-solar output scenes corresponding to the respective clusters, and determining the most representative representative wind-solar output scene as the final wind-solar typical scene.

[0096] Furthermore, after obtaining the occurrence probability and correlation coefficient of the representative wind-solar output scene corresponding to each cluster, this embodiment comprehensively considers these two factors to determine the most representative wind-solar typical scene. Exemplarily, the occurrence probability and the correlation coefficient can be weighted summed or multiplied to obtain a comprehensive evaluation index. Then, according to the size of this comprehensive evaluation index, the various representative wind-solar output scenes are sorted, and the scene with the highest evaluation index is selected as the final wind-solar typical scene. This embodiment can select the most representative one or several scenes from multiple possible scenes to provide reliable data support for the optimized operation of the power system.

[0097] For the overall implementation process of this embodiment, please refer to Figure 3 A flow chart of another typical wind-solar scene generation method is shown. In this embodiment, the historical wind-solar output data of the target wind-solar power station is first input, and the marginal distribution of wind-solar energy output is estimated, including:

[0098] (1) Based on the original wind and solar power output data, the probability density functions of wind power and photovoltaic power output are established for each period of 24 hours in a day.

[0099] Suppose the sample points extracted from the original wind and solar power output data X are X1, X2, ..., X n , f(x) is the probability density function that X meets, X∈R, this embodiment uses the kernel density estimation method (KDE) to establish the probability density function of wind power and photovoltaic output in each period of 24 hours in a day, where the kernel density function is a Gaussian kernel function. The probability density function is obtained by the following formula:

[0100]

[0101] Where n represents the number of sample points, m represents the number of variables, h represents the bandwidth, K(·) represents the kernel density function, and They represent the actual data of wind power output and photovoltaic power output in the i-th period of the d-th day respectively; i and i They represent the wind power output and photovoltaic output in the i-th period respectively.

[0102] (2) By integrating the probability density function, we can obtain the marginal distribution functions of wind power and photovoltaic power output F(X) and F(Y):

[0103]

[0104] Next, this embodiment constructs a joint distribution model of wind and solar power output based on a mixed Copula function, including:

[0105] (1) Realize the construction of the mixed Copula function. The expression of the mixed Copula function is as follows:

[0106] C M (F1(x1),F2(x2))=λ G C G (F1(x1),F2(x2);θ G )

[0107] +λ C C C (F1(x1),F2(x2);θ C )

[0108] +λ F C F (F1(x1),2(x2);θ F );

[0109]

[0110]

[0111]

[0112] Among them, C G , C C , C F Respectively represent GumbelCopula function, ClaytonCopula function and FrankCopula function, λ G ,λ C ,λ F Represents the weight parameter of the corresponding Copula function, θ G ,θ C ,θ F Represents the dependent parameter of the corresponding Copula function and satisfies 0≤λ G , C ,F ≤1,λ G +λ C +λ F =1,θ G ∈[1,+∞),θ C ∈[0,+∞),θ F ≠ 0. F1(x1) represents a further refined distribution function related to wind power output (x1 represents a variable related to wind power output), and F2(x2) represents a further refined distribution function related to photovoltaic output (x2 represents a variable related to photovoltaic output).

[0113] (2) Realize the weight parameter estimation of the mixed Copula function. The weight parameter estimation problem of the mixed Copula function can be regarded as a general optimization problem, and the least squares method is used to solve the value of the weight parameter.

[0114] 1) Objective function:

[0115] The objective function is to minimize the Euclidean distance between the hybrid Copula function and the empirical Copula function. The empirical Copula function is a reference function obtained based on actual data. By minimizing the Euclidean distance between the hybrid Copula function and the empirical Copula function, the hybrid Copula function can better fit the correlation relationship of the actual data. As shown below:

[0116]

[0117] 2) Constraints:

[0118] λ G +λ C +λ F =1;

[0119] 0≤λ G ≤1;

[0120] 0≤λ C ≤1;

[0121] 0≤λ F ≤1;

[0122] Among them, C e (u,v) represents a mixed Copula function, C M (u,v) represents the empirical Copula function, and u and v represent two variables related to the Copula function.

[0123] Afterwards, this embodiment samples and splices the joint distribution model of wind and solar output to generate a set of typical wind and solar sceneries, including:

[0124] (1) Based on the marginal distribution function of wind power output, the marginal distribution function of wind power output in each period of the night time is sampled according to the sampling number, and the output of the sampled wind points in each period of the night is obtained by inverse transformation of the joint probability distribution function of wind and solar power to generate the typical output curve of night wind power; for the night photovoltaic output, the zero value is taken as the typical output of night photovoltaic according to the sampling number.

[0125] (2) Based on the mixed Copula model, the sampling number is determined, and the joint probability distribution function of each time period during the day is sampled using the random sampling method. The wind power and photovoltaic output during the daytime period is obtained by inverting the joint probability distribution function.

[0126] (3) The typical outputs at different time periods during the day and night are combined to obtain a set of typical wind and solar output scenes for the entire day.

[0127] That is to say, for the night time period, this embodiment samples the wind power output edge distribution function of each time period in the night time period according to the sampling number based on the wind power output edge distribution function. The sampling number here is a predetermined value, which determines the number of wind power sampling points selected from the edge distribution function for the night time period. Through sampling, some possible values ​​of wind power output in different time periods at night can be obtained. The output of the sampled wind points in each time period at night is obtained by inverse transformation of the joint probability distribution function of wind and light. The joint probability distribution function contains the correlation information between the wind and light outputs. Through the inverse transformation operation, the points sampled from the edge distribution function can be converted into actual output values, thereby generating a typical output curve of wind power at night. This curve can reflect the typical output of wind power in different time periods at night. For photovoltaic output at night, since there is usually no light at night, the photovoltaic output is very little or zero. Therefore, this embodiment takes a zero value as the typical photovoltaic output at night according to the sampling number. For the daytime period, this embodiment determines the sampling number based on the mixed Copula model. The hybrid Copula model takes into account the correlation between wind and solar output. The sampling number determined by this model can better reflect the situation of wind and solar output during the day. Then, the random sampling method is used to sample the joint probability distribution function of each period during the day. The wind power and photovoltaic output during the daytime period is obtained by inverting the joint probability distribution function. The inversion operation converts the daytime wind and photovoltaic sampling points sampled from the joint probability distribution function into actual output values, so that the specific output of wind power and photovoltaic power at different periods during the day can be obtained. The typical output of each period during the day and night is spliced ​​together. The typical output curve of wind power at night, the typical output of photovoltaic power at night (zero value), and the typical output of wind power and photovoltaic power at different periods during the day are spliced ​​together in chronological order. Please refer to here. Figure 4The schematic diagram of typical output during the day and night is shown, and finally the scene set of typical wind and solar output for the whole day is obtained. This scene set contains the typical wind and solar output conditions at various times from night to day in a day, which can provide basic data for subsequent analysis and processing, such as evaluating the power generation capacity of wind and solar energy and analyzing its impact on the power system.

[0128] Then, this embodiment realizes the reduction of wind and solar power output scenarios based on the K-means clustering algorithm and the elbow rule, and calculates the probability of each scenario by the number of samples of the corresponding class of the cluster center, and uses the Euclidean distance to evaluate the distance between the fundamental point data and the cluster center:

[0129]

[0130] Among them, x represents the sample of each cluster, C i represents the cluster center of the i-th category, m represents the dimension of the sample data; X j represents the sample data x of the jth dimension; C ij Represents the cluster center C of the jth dimension i .

[0131] SSE is used as the evaluation index of the elbow method, and k represents the number of clusters:

[0132]

[0133] Finally, this embodiment selects the final typical scene of wind and solar based on the Pearson correlation coefficient and the probability of scene occurrence. This embodiment introduces the Pearson correlation coefficient to calculate the correlation between the historical wind power data and the typical wind scene obtained by clustering:

[0134]

[0135] Among them, r(X,Y) represents the Pearson correlation coefficient, which ranges from [-1, 1]. The closer r(X,Y) is to -1, the stronger the complementarity is, and the closer r(X,Y) is to 1, the stronger the correlation is. t , Y t Sequences X0 and Y0 in the time interval T are respectively the average values ​​of the output sample sequences of wind energy and solar energy.

[0136] The following is an explanation of the contents disclosed in the above embodiment through simple examples:

[0137] This example takes a new energy base that includes wind power and photovoltaic power as the research object, with a wind power installed capacity of 500,000 kW and a photovoltaic installed capacity of 1,000,000 kW. The historical output data of wind and photovoltaic power is hourly data for the whole year of 2022. Figure 5 and Figure 6A schematic diagram of a probability curve of the edge distribution of photovoltaic power output of a new energy base and a schematic diagram of a probability curve of the edge distribution of wind power output of a new energy base are shown respectively.

[0138] Please refer to Table 1 for the information comparison of each Copula function model in each time period. The Euclidean distance test of the constructed hybrid Copula model in this embodiment shows that the Euclidean distance of the hybrid Copula function model is always better than that of the single Copula function model, indicating that the hybrid Copula function model can better fit the correlation relationship between wind and solar output.

[0139] Table 1

[0140]

[0141]

[0142] Please refer to Table 2 for the occurrence probability and correlation coefficient of each scenario and Table 3 for the Pearson correlation coefficient of the original data (i.e., historical wind and solar output data). In this embodiment, the sampling number is determined to be 10,000. An initial set of typical wind and solar output scenarios is generated by random sampling. Based on the elbow method and K-means clustering algorithm, 8 groups of typical wind and solar scenarios are determined. The probability of occurrence of each scenario and the correlation relationship of wind and solar output are calculated. Please refer to Figure 7 The schematic diagram of the typical wind-solar scenario results obtained after K-means clustering processing shows that Scheme 1 (i.e., Scenario 1) is the most representative typical wind-solar output scenario.

[0143] Table 2

[0144]

[0145] Table 3

[0146]

[0147]

[0148] In summary, this embodiment is based on the marginal distribution function of wind and solar energy output, and uses a hybrid Copula function to construct a joint distribution model of wind and solar output. It is not limited to a single Copula function, but can adaptively adjust the model parameters according to the historical wind and solar output data. It can flexibly change the parameters in the hybrid Copula function according to the specific situation of the actual wind and solar output data, so that the constructed model can better fit the data characteristics, thereby generating a joint distribution model of wind and solar output with the best fitting effect. Since this embodiment can generate a wind and solar time series joint output model with the best fitting effect, and select the most representative wind and solar typical scene in combination with the probability of occurrence of typical scenes, it has higher accuracy in reflecting the actual wind and solar output in each region, and can more accurately capture the different characteristics and changing laws of wind and solar output caused by differences in geographical, climatic and other factors in different regions. In addition, the most representative wind and solar typical scenes selected by this embodiment and the related probability and correlation coefficient data can provide reliable data support for the planning, scheduling, risk assessment and other aspects of the power system.

[0149] In this embodiment, a scenery typical scene generation device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0150] This embodiment provides a typical scene generation device. Figure 8 As shown, including:

[0151] The marginal distribution function acquisition module 801 is used to acquire the historical wind and solar power output data of the target wind and solar power station, and calculate the wind and solar energy output marginal distribution function of the target wind and solar power station based on the historical wind and solar power output data;

[0152] A joint distribution model building module 802 is used to build a joint distribution model of wind and solar power output using a mixed Copula function based on the wind and solar power output marginal distribution function;

[0153] A wind-solar typical scene set generation module 803 is used to perform sampling and splicing processing on the wind-solar output joint distribution model to generate a wind-solar typical scene set;

[0154] The final typical scene generation module 804 is used to calculate the occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical output data of the wind power, and select the final typical scene based on the occurrence probability and the correlation coefficient.

[0155] In some optional implementations, the marginal distribution function acquisition module 801 is further used to:

[0156] According to the historical wind and solar power output data, a probability density function of each time period in the target period is established respectively; the probability density function includes the wind power probability density function and the photovoltaic output probability density function;

[0157] The probability density function is integrated to obtain the wind-solar energy output marginal distribution function of the target wind-solar power station; the wind-solar energy output marginal distribution function includes the wind power marginal distribution function and the photovoltaic marginal distribution function.

[0158] In some optional implementations, the joint distribution model building module 802 is further used to:

[0159] Obtain weight parameters and dependent parameters corresponding to each function in the hybrid Copula function to be constructed; the hybrid Copula function includes a Gumbel Copula function, a Clayton Copula function and a Frank Copula function;

[0160] Based on the weight parameters and dependent parameters corresponding to each function, construct the mixed Copula function;

[0161] The wind-solar energy output marginal distribution function and the mixed Copula function are combined to construct the wind-solar output joint distribution model.

[0162] In some optional implementations, the scenery typical scene set generation module 803 is further used to:

[0163] Based on the joint distribution model of wind and solar power output, a wind and solar power joint probability distribution function is obtained;

[0164] Based on the wind power output marginal distribution function, each night time period is sampled to obtain the wind power sampling point in each night time period;

[0165] For the wind power sampling points in the night time period, the wind-solar joint probability distribution function is used to perform an inverse transformation to obtain the typical output value of night wind power in each night time period;

[0166] Based on the mixed Copula function, the wind-solar joint probability distribution function in each daytime period is sampled to obtain the wind power and photovoltaic sampling points in each daytime period;

[0167] For the wind power and photovoltaic sampling points during the daytime period, the typical output value of daytime wind power and the typical output value of daytime photovoltaic are generated by inverting the wind-solar joint probability distribution function;

[0168] The typical wind power output value at night, the typical wind power output value during the day and the typical photovoltaic output value during the day are spliced ​​together to generate a typical wind and solar scene set for the whole day.

[0169] In some optional implementations, the final scenery typical scene generation module 804 is further used to:

[0170] Obtaining an optimal number of clusters, and clustering the typical wind and solar scene set based on the optimal number of clusters to obtain each cluster, as well as the number of samples and cluster centers of each cluster; the cluster centers indicate representative wind and solar output scenes corresponding to each cluster;

[0171] For each cluster, based on the number of samples of the cluster and the total number of samples of the typical scene set of wind and solar power, the occurrence probability of the representative wind and solar power output scene corresponding to each cluster is obtained;

[0172] Calculate the correlation coefficient between the representative wind and solar power output scene corresponding to each cluster and the wind and solar power historical output data;

[0173] The representativeness of the representative wind-solar output scenes corresponding to each cluster is obtained by combining the occurrence probability with the correlation coefficient, and the representative wind-solar output scene with the highest representativeness is determined as the final wind-solar typical scene.

[0174] In some optional implementations, the final scenery typical scene generation module 804 is further used to:

[0175] Obtaining a value range of the number of clusters, performing clustering processing on the typical scenery scene set based on the value range of the number of clusters, and calculating the sum of squared errors corresponding to each number of clusters in the value range;

[0176] A relationship curve between each cluster number in the value range and the error sum corresponding to the cluster number is constructed, and a target downward trend turning point of the error sum in the relationship curve is determined as the optimal cluster number.

[0177] In summary, this embodiment is based on the marginal distribution function of wind and solar energy output, and uses a hybrid Copula function to construct a joint distribution model of wind and solar output. It is not limited to a single Copula function, but can adaptively adjust the model parameters according to the historical wind and solar output data. It can flexibly change the parameters in the hybrid Copula function according to the specific situation of the actual wind and solar output data, so that the constructed model can better fit the data characteristics, thereby generating a joint distribution model of wind and solar output with the best fitting effect. Since this embodiment can generate a wind and solar time series joint output model with the best fitting effect, and select the most representative wind and solar typical scene in combination with the probability of occurrence of typical scenes, it has higher accuracy in reflecting the actual wind and solar output in each region, and can more accurately capture the different characteristics and changing laws of wind and solar output caused by differences in geographical, climatic and other factors in different regions. In addition, the most representative wind and solar typical scenes selected by this embodiment and the related probability and correlation coefficient data can provide reliable data support for the planning, scheduling, risk assessment and other aspects of the power system.

[0178] See also Fig. 9 , Fig. 9 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig. 9 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig. 9 A processor 10 is taken as an example.

[0179] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0180] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0181] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0182] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0183] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0184] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0185] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0186] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the defined scope.

Claims

1. A method for generating typical scenery scenes, characterized in that: The method comprises: Acquire historical wind and solar power output data of a target wind and solar power station, and calculate a wind and solar energy output marginal distribution function of the target wind and solar power station based on the historical wind and solar power output data; Based on the wind and solar energy output marginal distribution function, a hybrid Copula function is used to construct a joint distribution model of wind and solar output; Sampling and splicing the joint distribution model of wind and solar power output to generate a set of typical wind and solar sceneries; The occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind and solar power output data are calculated, and a final typical scene is selected based on the occurrence probability and the correlation coefficient.

2. The method according to claim 1, characterized in that The calculating of the wind and solar energy output marginal distribution function of the target wind and solar power station based on the wind and solar historical output data includes: For the historical wind and solar power output data, a probability density function is established for each time period in the target period; the probability density function includes a wind power probability density function and a photovoltaic output probability density function; The probability density function is integrated to obtain the wind-solar energy output marginal distribution function of the target wind-solar power station; the wind-solar energy output marginal distribution function includes a wind power marginal distribution function and a photovoltaic marginal distribution function.

3. The method according to claim 1, characterized in that Based on the wind and solar energy output marginal distribution function, a hybrid Copula function is used to construct a joint distribution model of wind and solar output, including: Obtaining weight parameters and dependent parameters corresponding to each function in the hybrid Copula function to be constructed; the hybrid Copula function includes a Gumbel Copula function, a Clayton Copula function and a Frank Copula function; Constructing the hybrid Copula function based on the weight parameters and dependent parameters corresponding to the functions; The wind-solar energy output marginal distribution function and the mixed Copula function are combined to construct the wind-solar output joint distribution model.

4. The method according to claim 2, characterized in that The sampling and splicing processing of the joint distribution model of the wind and solar power output to generate a set of typical wind and solar sceneries includes: Based on the wind-solar power joint distribution model, a wind-solar power joint probability distribution function is obtained; Sampling each night time period based on the wind power output marginal distribution function to obtain wind power sampling points in each night time period; For the wind power sampling points in the night time period, inverse transformation is performed using the wind-solar joint probability distribution function to obtain typical output values ​​of night wind power in each night time period; Based on the hybrid Copula function, the wind-solar joint probability distribution function of each daytime period is sampled to obtain wind power and photovoltaic sampling points of each daytime period; For the wind power and photovoltaic sampling points during the daytime period, the typical output value of the daytime wind power and the typical output value of the daytime photovoltaic are generated by inverting the wind-solar joint probability distribution function; The typical output values ​​of nighttime wind power, the typical output values ​​of daytime wind power and the typical output values ​​of daytime photovoltaic power in each period of daytime and nighttime are spliced ​​to generate a typical scene set of wind and solar power for the whole day.

5. The method according to any one of claims 1 to 4, characterized in that: The calculating the occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind power output data, and selecting the final typical scene based on the occurrence probability and the correlation coefficient, includes: Obtaining an optimal number of clusters, and clustering the typical wind-solar scene set based on the optimal number of clusters to obtain each cluster, as well as the number of samples and cluster centers of each cluster; the cluster centers indicate representative wind-solar output scenes corresponding to each cluster; For each of the clusters, based on the number of samples of the cluster and the total number of samples of the typical scene set of wind and solar power, the occurrence probability of the representative wind and solar power output scene corresponding to each cluster is obtained; Calculating the correlation coefficient between the representative wind / solar power output scene corresponding to each cluster and the wind / solar power historical output data; In combination with the occurrence probability and the correlation coefficient, the representativeness of the representative wind-solar output scenes corresponding to the respective clusters is obtained, and the representative wind-solar output scene with the highest representativeness is determined as the final wind-solar typical scene.

6. The method according to claim 5, characterized in that The obtaining of the optimal number of clusters comprises: Acquire a value range of the number of clusters, perform clustering processing on the typical scene set based on the value range of the number of clusters, and calculate the sum of squared errors corresponding to each number of clusters in the value range; A relationship curve between each cluster number in the value range and the error sum corresponding to the cluster number is constructed, and a target downward trend turning point of the error sum in the relationship curve is determined as the optimal cluster number.

7. A typical scene generation device, characterized in that: The device comprises: An edge distribution function acquisition module is used to acquire the wind and solar historical output data of the target wind and solar power station, and calculate the wind and solar energy output edge distribution function of the target wind and solar power station based on the wind and solar historical output data; A joint distribution model building module, used to build a joint distribution model of wind and solar power output using a mixed Copula function based on the wind and solar power output marginal distribution function; A wind-solar typical scene set generation module, used for sampling and splicing the wind-solar joint distribution model to generate a wind-solar typical scene set; The final typical scene generation module is used to calculate the occurrence probability of each target scene in the typical scene set and the correlation coefficient with the historical wind power output data, and select the final typical scene based on the occurrence probability and the correlation coefficient.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes a typical scenery scene generation method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute a typical scenery scene generation method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions are used to enable a computer to execute a typical scenery scene generation method according to any one of claims 1 to 6.

Images