A new energy output power random generation method based on mixed distribution function and high-dimensional vine copula function

Abstract

The application discloses a new energy output power random generation method based on a hybrid distribution function and a high-dimensional vine Copula function, and comprises the following steps: collecting wind and light natural resource data and engineering data, calculating wind and light output power at an hourly scale, and establishing a wind and light output characteristic description index system; defining a discrete-continuous hybrid marginal distribution function to describe the intermittence characteristics of the wind and light output power at the hourly scale; constructing a high-dimensional hybrid vine Copula function with a self-defined structure to describe the correlation characteristics between the wind and light output power at each hour within a day; and based on the discrete-continuous hybrid marginal distribution function and the high-dimensional vine Copula joint distribution function with the self-defined structure, randomly simulating and generating a wind and light output power scenario set within a day. The application can accurately simulate the wind and light output power value at the hourly scale and retain the intermittence and correlation characteristics, and is of great significance for planning a new power system operation mainly based on new energy.

Description

Technical Field

[0001] This invention belongs to the field of multi-energy complementary power generation dispatch technology, and in particular, it is a method for random generation of new energy output power based on a hybrid distribution function and a high-dimensional Copula function. Background Technology

[0002] At present, the installed capacity and grid-connected electricity of new energy sources continue to rise. Due to the inherent intermittent and random fluctuation characteristics of these energy sources, the large-scale access of new energy sources poses a severe challenge to the security, stability and power quality of the power grid. Research on the characteristics of renewable energy generation and simulation modeling are of great value for improving the operational stability of new power systems.

[0003] Currently, methods for simulating renewable energy power output typically include physical modeling, machine learning, and mathematical statistics. Physical modeling requires high-resolution spatiotemporal natural resource datasets and suffers from high computational complexity; machine learning demands a large number of data samples and has poor generalization ability; mathematical statistics, on the other hand, is characterized by its simple structure and high reliability, enabling rapid modeling of renewable energy power output. However, traditional mathematical statistics modeling methods, such as Monte Carlo and Copula functions, struggle to effectively characterize the intermittent nature of renewable energy output across its dual boundaries and its intraday higher-order temporal histories, ultimately reducing the accuracy of renewable energy power output simulations. Summary of the Invention

[0004] Purpose of the invention: The purpose of this invention is to provide a method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function. This method can accurately simulate the hourly scale of wind and solar power output while preserving its intermittent and correlation characteristics, which is of great significance for planning the operation of a new type of power system based on new energy sources.

[0005] Technical solution: The method described in this invention includes the following steps:

[0006] S1. Based on the hourly wind and solar natural resource data and engineering data of the new energy base, calculate the hourly wind and solar power output samples of the base as observation samples, and establish a descriptive index system for wind and solar power output characteristics.

[0007] S2. Construct a hybrid marginal distribution function based on the Dirac and Kernal functions, divide the daily hourly wind and solar power output observation samples into discrete and continuous parts, and fit the power output distribution of the discrete and continuous parts based on the Dirac and Kernal functions in the hybrid marginal distribution function.

[0008] S3. A high-dimensional hybrid vine Copula is used to connect the hybrid marginal distribution function. The corresponding vine structure, Copula function type and parameters are selected. The goodness-of-fit evaluation index is used to evaluate the fitting effect of the distribution function, thereby constructing a custom structure high-dimensional hybrid vine Copula joint distribution function that describes the correlation characteristics between the output power of each hour in the day.

[0009] S4. Based on the constructed custom structure high-dimensional hybrid vine Copula joint distribution function, the wind and solar power output for each hour of the day is derived by inverse sampling, thereby randomly generating a set of wind and solar power output scenarios that retain intermittency and correlation characteristics as simulation samples.

[0010] S5. Calculate the evaluation index values ​​of the observed samples and simulated samples based on the descriptive index system of wind and solar power output characteristics, and compare the evaluation index values ​​of the observed samples and the evaluation index values ​​of the simulated samples to evaluate the descriptive effect of the method on the randomness, intermittency and correlation characteristics of wind and solar power output at the hourly scale.

[0011] Furthermore, step S1 includes the following steps:

[0012] S11. Collect hourly wind and solar natural resource data and engineering data of the new energy base. The wind and solar natural resource data includes hourly wind speed, air density, altitude, solar radiation and temperature. The engineering data includes the number of wind turbines, installed capacity, output coefficient, cut-in wind speed, rated wind speed, cut-out wind speed, hub area, hub height, wind speed conversion coefficient, number of photovoltaic panels, installed capacity, area and heat loss efficiency coefficient.

[0013] S12. Calculate the hourly wind and solar power output of the new energy base based on the calculation model of a single wind turbine and photovoltaic power output, wind and solar natural resource data and engineering data;

[0014] S13. Establish a descriptive index system for the randomness, intermittency, and correlation characteristics of wind and solar power output at the hourly scale; including:

[0015] (1) The randomness characteristics of wind and solar power output on an hourly scale are measured by the sample mean, standard deviation, coefficient of variation, kurtosis coefficient and power output volatility.

[0016] (2) The power output activity is used to measure the intermittent characteristics of wind and solar power output on an hourly scale;

[0017] (3) Kendall correlation coefficient is used to measure the temporal correlation characteristics between wind and solar power output on an hourly scale.

[0018] Furthermore, step S2 specifically includes:

[0019] S21. Compile the hourly output power of wind power and photovoltaic power into a sample group with a length of [missing information]. Dimensions The sample spaces are respectively 24-dimensional random variables ;

[0020] S22. Considering the upper and lower boundary properties of wind and solar power output at the hourly scale, the Dirac function and the Kernal function are introduced to represent the upper and lower boundary values ​​and continuous values ​​in wind and solar power output, respectively. A hybrid marginal distribution function is constructed to describe the double boundary distribution characteristics of wind and solar power output variables at the hourly scale.

[0021] S23. Use the root mean square error (RMSE) and goodness-of-fit R². 2 The evaluation criteria assess the fitting effect of the hybrid marginal distribution function constructed in step S22 on the distribution of the dual-boundary wind and solar power output variables at the hourly scale, which includes a large number of zero values ​​and rated output values.

[0022] Furthermore, step S22 specifically includes:

[0023] (1) Construct a mixed probability density function for the hour-scale double-boundary wind and solar power output variables, expressed as:

[0024] ;

[0025] ;

[0026] in, Let be the probability density function of wind power output; Let be the probability density function of the photoelectric output power; express The Dirac function at that location, and They represent and The Dirac function at the location; These represent the Kernal functions obtained by fitting the power output variables of wind power or solar power, respectively. For random variables The value in the random variable For a certain period of time The output power of wind power or solar power; This is the intermittency coefficient;

[0027] (2) The intermittency coefficient is determined by the proportion of the number of days with zero wind or solar power output and the number of days with rated power in a certain period to the total number of days in that period; the expression is:

[0028] ;

[0029] (3) The final cumulative distribution function, i.e., the mixed marginal distribution function, is derived from the constructed mixed probability density function. :

[0030] ;

[0031] ;

[0032] in, Here is the definition of the cumulative distribution function, and is a random variable. Less than or equal to The probability of; Represented as When, random variable Less than or equal to The probability of.

[0033] Furthermore, the Dirac function The probability density used to represent the discrete part is expressed as:

[0034] ;

[0035] ;

[0036] Kernal function The probability density used to represent the continuous portion is expressed as:

[0037] ;

[0038] ;

[0039] in, For a certain period of time Number of wind / solar power output samples; Indicates the first Individual wind power / solar power output values; For kernel function parameters, The optimal parameters; For random variables The standard deviation.

[0040] Furthermore, step S3 specifically includes:

[0041] (1) Treat the hourly wind and solar power output variables as nodes and calculate the Kendall correlation coefficients between all nodes in the first-level tree structure and the remaining nodes in turn. ;

[0042] (2) Estimate the Copula function type and parameter value connecting two nodes in the first-level tree structure, and use the AIC criterion to determine the optimal Copula function type from the aspects of goodness of fit and complexity; ,in, and These represent the number of parameters of the currently selected Copula function and the value of the maximum likelihood function of the Copula function, respectively.

[0043] (3) The parameters of the Copula function are determined by the maximum likelihood estimation method;

[0044] (4) After determining the first-level Copula function, sample the Copula function of each side in turn to obtain the corresponding node value of the next level tree. Repeat the above steps (1) to (3) to determine the tree structure and Copula function type of the next level until the tree structure and Copula function type of all levels are determined.

[0045] (5) Based on the difference between the Cramer-von Mises statistical empirical Copula function and the constructed Copula function, the Copula model with the best fitting effect is selected as: ,in, To test the goodness of fit between empirical Copula functions and constructed Copula functions, and Let represent the empirical Copula distribution function and the constructed Copula distribution function, respectively.

[0046] Furthermore, step S4 specifically includes:

[0047] (1) Derive hourly scale wind and solar power output samples The corresponding marginal distribution function, joint distribution function, and joint conditional distribution function;

[0048] (2) Generating uniformly distributed data based on the Monte Carlo method 3D random vector The distribution function of the wind and solar power output sequence for each time period is solved iteratively.

[0049] ;

[0050] in, Corresponding representation The value in; For the generated random variables that follow a uniform distribution The inverse distribution function, All of these are the conditional distribution functions in the joint distribution function of the high-dimensional hybrid vine Copula, which is a custom structure constructed by the user.

[0051] (3) Based on the formula in step (2), derive the power distribution function for each time period by finding the inverse function; then the first... Power distribution function during time period for;

[0052] ;

[0053] in, for joint conditional distribution function The inverse function of the concise expression of .

[0054] (4) Repeat the sampling process of steps (2) to (3) in each layer of the subsequent vine Copula structure, and derive the distribution function of the nodes associated with each time period layer by layer according to the optimal vine Copula structure, function type and parameters, and finally obtain the distribution function of wind and solar power output for each hour containing complete correlation.

[0055] (5) By substituting the randomly generated hourly wind and solar power output distribution function into the inverse function of the fitted hourly cumulative wind and solar power output distribution function, the hourly wind and solar power output samples generated by random simulation are obtained, i.e., the simulation samples.

[0056] The system corresponding to the method includes:

[0057] The data and indicator system unit is used to calculate the wind and solar power output samples at the hourly scale of the new energy base based on the wind and solar natural resource data and engineering data of the base, and to establish a descriptive indicator system for the characteristics of wind and solar power output.

[0058] The hybrid marginal distribution function construction unit is used to construct a hybrid marginal distribution function based on the Dirac function and the Kernal function. It divides the wind and solar power output observation samples of each hour of the day into discrete and continuous parts, and fits the power output distribution of the discrete and continuous parts based on the Dirac function and the Kernal function in the hybrid marginal distribution function.

[0059] The joint distribution function construction unit is used to connect the hybrid marginal distribution function with a high-dimensional hybrid vine Copula, optimize the corresponding vine structure, Copula function type and parameters, and evaluate the fitting effect of the distribution function with the goodness-of-fit evaluation index, thereby constructing a custom structure high-dimensional hybrid vine Copula joint distribution function that describes the correlation characteristics between the output power of each hour in the day.

[0060] The wind and solar power output simulation unit is used to derive the wind and solar power output for each hour of the day by inverse sampling based on the constructed custom structure high-dimensional hybrid vine Copula joint distribution function, thereby randomly generating a set of wind and solar power output scenarios that retain intermittency and correlation characteristics as simulation samples.

[0061] The evaluation unit is used to calculate the evaluation index values ​​of the observed samples and the simulated samples based on the descriptive index system of wind and solar power output characteristics, and compare the evaluation index values ​​of the observed samples and the evaluation index values ​​of the simulated samples to evaluate the descriptive effect of the method on the randomness, intermittency and correlation characteristics of wind and solar power output at the hourly scale.

[0062] The electronic device of the present invention includes a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor. When the computer program / instruction is executed by the processor, it implements the steps of the method for random generation of new energy power output based on a hybrid distribution function and a high-dimensional Copula function.

[0063] The computer-readable storage medium of the present invention stores computer instructions, which, when invoked, are used to execute the steps of the method for random generation of new energy power output based on a hybrid distribution function and a high-dimensional Copula function.

[0064] Beneficial effects: Compared with the prior art, the significant technical effects of this invention are as follows: Considering the inherent intermittency and high-order time history correlation characteristics of wind and solar new energy, a "discrete-continuous" hybrid marginal distribution function is introduced to solve the problem of double boundary distribution and intermittency of wind and solar power output; a high-dimensional Copula model is constructed and combined with the "discrete-continuous" hybrid marginal distribution function to characterize the complex correlation structure and high-order time history correlation of wind and solar power output, accurately simulating the hourly-scale power output value of wind and solar while retaining its output characteristics, thereby improving the operating efficiency of the new power system with new energy as the main body. Attached Figure Description

[0065] Figure 1 This is a flowchart of the method of the present invention;

[0066] Figure 2 These are comparison charts showing the improvement effects of fitting the hybrid marginal distribution function to the hourly scale wind and solar power output distribution characteristics. (a) shows the improvement effects of fitting the wind power output distribution characteristics, and (b) shows the improvement effects of fitting the solar power output distribution characteristics.

[0067] Figure 3 This is a schematic diagram of the optimal 24-dimensional Copula structure;

[0068] Figure 4These are comparison images of randomly generated simulated wind and solar power output samples and observed samples. Among them, (a) is a comparison image of simulated wind power output samples and observed samples, and (b) is a comparison image of simulated solar power output samples and observed samples. Detailed Implementation

[0069] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. However, it will be apparent to those skilled in the art that the present invention can be practiced without one or more of these details. In other instances, some technical features well-known in the art have not been described to avoid confusion with the present invention. The order of the relevant steps in the present invention is not limiting; that is, those skilled in the art can adjust it. The order in the present invention is an exemplary writing style, not a limiting description.

[0070] This invention discloses a stochastic simulation method for renewable energy output using a hybrid distribution function and a high-dimensional Copula function. This method constructs a model coupling a "discrete-continuous" hybrid marginal distribution function and a high-dimensional Copula joint distribution function. This model can accurately simulate hourly-scale wind and solar power output while preserving its randomness, intermittency, and correlation characteristics, thus improving the operational efficiency of new power systems primarily based on renewable energy. Compared to traditional single-type marginal distribution functions and low-dimensional Copula functions, this method addresses the problems of inaccurate characterization of intermittent distribution characteristics and correlation structures, achieving accurate simulation of hourly-scale wind and solar power output while retaining its intermittency and correlation characteristics. This is of great significance for planning the operation of new power systems primarily based on renewable energy.

[0071] like Figure 1 As shown, the method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function, as described in this invention, includes the following steps:

[0072] S1. Collect hourly wind and solar natural resource data and engineering data of new energy bases, calculate hourly wind and solar power output samples (i.e. observation samples) of the bases, and establish a descriptive index system for wind and solar power output characteristics to characterize the randomness, intermittency and correlation of wind and solar power output.

[0073] According to one aspect of this application, step S1 is further comprising:

[0074] S11. Collect hourly-scale wind and solar natural resource data and engineering data of the new energy base. The wind and solar natural resource data includes hourly-scale wind speed, air density, altitude, solar radiation and temperature. The engineering data includes the number of wind turbines, installed capacity, output coefficient, cut-in wind speed, rated wind speed, cut-out wind speed, hub area, hub height, wind speed conversion coefficient, number of photovoltaic panels, installed capacity, area and heat loss efficiency coefficient.

[0075] In this embodiment, the aforementioned wind and solar natural resource data can be obtained from publicly available data. The purpose is to gain a deeper understanding of the basic conditions of the study area and the characteristics of wind and solar power output, so as to provide basic information and data support for subsequent model design and construction.

[0076] S12. Calculate the hourly wind and solar power output of the new energy base based on the calculation model of a single wind turbine and photovoltaic power output, wind and solar natural resource data and engineering data.

[0077] According to one aspect of this application, step S12 further comprises:

[0078] S12a. Based on the power output calculation model of a single wind turbine, wind resource data, and engineering data of the wind turbine, the hourly wind power output value is calculated:

[0079] (1)

[0080] in, The output of a single wind turbine during a time period t is expressed in kW. Rated output of the wind turbine, in kW; This is the output coefficient of the wind turbine generator; The area of ​​the wind turbine hub is m. 3 ; air density, kg / m³ 3 ; , and These are the cut-in, rated, and cut-out wind speeds of the wind turbine, respectively, in m / s, corresponding to the critical wind speeds at which the wind turbine starts generating electricity, operates at its rated state, and stops generating electricity. It is the wind speed at the hub of the wind turbine, in m / s;

[0081] Considering the mismatch between the height corresponding to wind speed data and the height of wind turbine generators, a wind speed conversion relationship can be established to convert the wind speed data to the height of the wind turbine generators. Approximate conversion to wheel hub elevation Wind speed:

[0082] (2)

[0083] in, , These represent the hub heights of different wind turbine generators, in meters (m). , They are respectively , Corresponding wind speed, m / s; The wind speed conversion coefficient is... , Altitude variation.

[0084] S12b. Based on the single photovoltaic power output calculation model, photovoltaic resource data, and photovoltaic panel engineering data, the hourly photovoltaic power output value is calculated:

[0085] (3)

[0086] in, For photovoltaic power station time period Output, kW; Under standard conditions (corresponding to solar radiation intensity) =1000W / m 2 ,temperature Photovoltaic panel output at 25℃ (kW); For photovoltaic power station time period Actual solar radiation intensity, W / m 2 ; The heat loss efficiency coefficient is expressed as % / ℃ (generally 0.45% / ℃ for monocrystalline silicon photovoltaic cells). For time period Temperature, °C; The area of ​​the photovoltaic panels in the photovoltaic power station is in meters. 2 .

[0087] S12c: Based on the power output calculation models of a single wind power station and photovoltaic power station in steps S12a and S12b, multiply by the number of wind power and photovoltaic power stations in the new energy base to obtain the hourly wind and solar power output samples (i.e., observation samples) of the base.

[0088] In this embodiment, since wind and solar energy are significantly and rapidly affected by external environmental conditions such as weather and topography, they exhibit high uncertainty, significant time-varying characteristics, and intermittent power output. Therefore, it is necessary to establish a comprehensive feature description index system to accurately describe the fluctuations and related characteristics of the power output of wind and solar power stations in adjacent time periods.

[0089] S13. Establish a descriptive index system for the randomness, intermittency, and correlation characteristics of wind and solar power output at the hourly scale.

[0090] According to one aspect of this application, step S13 further comprises:

[0091] S13a. The randomness of wind and solar power output on an hourly scale is measured using the sample mean, standard deviation, coefficient of variation, kurtosis coefficient, and output volatility. The formula for calculating output volatility is:

[0092] (4)

[0093] in, For the current time period The fluctuation rate of wind or solar power output; For the current time period The average output power of wind power or photovoltaic power, in kW; for Average wind or solar power output during the time period, in kW; The installed capacity is wind power or photovoltaic power, in kW.

[0094] S13b. Using power output activity to measure the intermittent characteristics of wind and solar power output on an hourly scale:

[0095] (5)

[0096] in, To contribute to activity level; This indicates the number of days during which the effort output is not zero. This indicates the total number of days in that period; Indicates the first Each wind / solar power output value ; This represents the total number of samples for wind and solar power output at the hourly scale.

[0097] S13c. Kendall correlation coefficient is used to measure the temporal correlation characteristics between wind and solar power output at the hourly scale:

[0098] (6)

[0099] in, Kendall correlation coefficient; These represent the number of pairs of samples with the same or different serial numbers after sorting by numerical value.

[0100] In this embodiment, the hourly wind and solar power output exhibits a significant double-boundary intermittent characteristic. In wind power systems, when the wind speed is below the cut-in wind speed threshold of the wind turbine, the wind power output is zero; when the wind speed exceeds the cut-out wind speed threshold, the wind power output reaches the rated power. In photovoltaic power systems, due to the influence of the diurnal cycle, power output is generated during the day and zero at night. Furthermore, the intermittency of photovoltaic power output varies with seasonal changes, and the power output sequence is discontinuous. By constructing a "discrete-continuous" hybrid marginal distribution function to describe the double-boundary distribution characteristics of hourly wind and solar power output samples containing a large number of zero values ​​and rated output values, compared with a uniformly modeled marginal distribution function, this invention uses the Dirac function to represent the discrete part of the wind and solar power output distribution and the Kernal function to represent the continuous part, thus solving the problem of inaccurate characterization of distribution features in a single-type marginal distribution function.

[0101] S2. Construct a "discrete-continuous" hybrid marginal distribution function based on the Dirac and Kernal functions. Divide the daily hourly wind and solar power output observation samples into discrete and continuous parts. Fit the power output distribution of the discrete and continuous parts based on the Dirac and Kernal functions in the hybrid marginal distribution function, thereby describing the intermittent characteristics of wind and solar power output at the hourly scale.

[0102] According to one aspect of this application, step S2 further comprises:

[0103] S21. Compile the hourly output power of wind power and photovoltaic power into a sample group with a length of (i.e., total number of days). Dimensions (i.e., within 24 hours), the sample space is respectively 24-dimensional random variables :

[0104] (7)

[0105] in, The installed capacity of wind power (i.e., the maximum output of wind power) is kW; The installed capacity of the photovoltaic system (i.e., the maximum output of the photovoltaic system) is in kW. These represent the observed samples consisting of wind power and solar power output for each hour of the day. They represent the first sky Wind and solar power output at any given time, in kW. , .

[0106] S22. Considering the upper and lower boundary characteristics of wind and solar power output at the hourly scale, the Dirac and Kernal functions are introduced to represent the upper and lower boundary values ​​and continuous values ​​in wind and solar power output, respectively. A "discrete-continuous" hybrid marginal distribution function is constructed to describe the double-boundary distribution characteristics of the hourly wind and solar power output variable. For example... Figure 2 Figures (a) and (b) demonstrate the improved fitting effect of the "discrete-continuous" hybrid marginal distribution function constructed using this step on the hourly wind and solar power output distribution characteristics.

[0107] According to one aspect of this application, step S22 further comprises:

[0108] S22a. Construct a hybrid probability density function that can characterize hourly-scale double-boundary wind and solar power output variables containing a large number of zero values ​​and rated output values. For a given time period... The wind / solar power output is denoted as a random variable. Their probability density functions can be written as follows:

[0109] (8)

[0110] (9)

[0111] in, Let be the probability density function of wind power output; Let be the probability density function of the photoelectric output power; express The Dirac function at that location is described in step S22b. and They represent and The Dirac function at the location; These represent the Kernal functions obtained by fitting the wind / solar power output variables, respectively. The calculation formula is shown in step S22c. For random variables (i.e., a certain period of time) The value in the wind power / solar power output; For the intermittency coefficient, see step S22d.

[0112] S22b, Step S22a and Dirac function in The probability density used to represent the discrete part:

[0113] (10)

[0114] (11)

[0115] S22c, Step S22a and Kernal function in (Right now ) is used to represent the probability density of the continuous portion:

[0116] (12)

[0117] (13)

[0118] in, For a certain period of time Number of wind / solar power output samples (i.e., total number of days); Indicates the first Individual wind power / solar power output values; Kernel function parameters, bandwidth, optimal parameters Calculated using empirical formulas; For random variables The standard deviation.

[0119] S22d, intermittency coefficient Use a certain time period respectively The ratio of days with zero wind / solar power output to days with rated power output to the total number of days in that period is used to determine the output:

[0120] (14)

[0121] in, They represent The proportion of days with zero wind power output and days with rated power output to the total number of days; express The proportion of days with zero photovoltaic power output and days with rated power output to the total number of days.

[0122] S22e, the random variable constructed from step S22a Mixed probability density function The cumulative distribution function of wind and solar power output on an hourly scale (i.e., the "discrete-continuous" hybrid marginal distribution function) can be derived separately. Then the corresponding variable Distribution function:

[0123] (15)

[0124] (16)

[0125] in, These are the cumulative power distribution functions for wind power and solar power output, respectively. The definition and derivation expression of the random variable is given. Less than or equal to The probability of;

[0126] The derivation expressions for the cumulative distribution functions when the power output of wind power and solar power is greater than 0 are respectively given. random variables Less than or equal to The probability of.

[0127] , These are the hybrid marginal distribution functions of wind power output and photovoltaic power output, respectively.

[0128] S23. Use the root mean square error (RMSE) and goodness-of-fit R². 2 The evaluation criteria assess the fitting effect of the "discrete-continuous" hybrid marginal distribution function constructed in step S22 on the distribution of the dual-boundary wind and solar power output variables at the hourly scale, which includes a large number of zero values ​​and rated output values.

[0129] In this embodiment, a custom-structured high-dimensional hybrid vine Copula joint distribution function is used to describe the correlation between the power output of each hour within a day. Compared with a low-dimensional, unstructured Copula family, the custom-structured high-dimensional hybrid vine Copula joint distribution function used in this invention can efficiently describe the complex correlation and structure between high-dimensional variables.

[0130] S3. Using a customizable high-dimensional hybrid vine Copula to connect the "discrete-continuous" hybrid marginal distribution function (i.e., the cumulative distribution function of wind and solar power output for each hour of the day) constructed in step S2, the corresponding vine structure, Copula function type and parameters are selected, and the goodness-of-fit evaluation index is used to evaluate the fitting effect of the distribution function, thereby constructing a customizable high-dimensional hybrid vine Copula joint distribution function that accurately describes the correlation characteristics between power output for each hour of the day.

[0131] According to one aspect of this application, step S3 further comprises:

[0132] S31. Derive the cumulative distribution function of wind and solar power output for 24 hours in a day (i.e., the "discrete-continuous" hybrid marginal distribution function) one by one according to step S22e.

[0133] S32. A high-dimensional hybrid vine Copula joint distribution function with a customizable structure is used to connect the cumulative distribution function of the 24-hour wind and solar power output within the day, and the corresponding optimal vine structure, Copula function type and parameters are selected.

[0134] According to one aspect of this application, step S32 further comprises:

[0135] S32a. Treating the hourly wind and solar power output variables as nodes, calculate the Kendall correlation coefficients between all nodes and the remaining nodes in the first-level tree structure. (The calculation method is the same as step S13c);

[0136] S32b. Estimate the Copula function type and parameter values ​​connecting two nodes in the first-level tree structure, and determine the optimal Copula function type using the AIC criterion from the perspectives of goodness of fit and complexity:

[0137] (17)

[0138] in, and These represent the number of parameters of the currently selected Copula function and the value of the maximum likelihood function of the Copula function, respectively.

[0139] S32c. Determine the parameters of the Copula function using the maximum likelihood estimation method:

[0140] (18)

[0141] in, This represents the likelihood function (i.e., the currently selected Copula function type). Indicates the given parameters The probability of each variable;

[0142] S32d: After determining the first-level Copula function, sample the corresponding node values ​​of the next-level tree for each edge Copula function in sequence. Repeat steps S32a~S32c to determine the next-level tree structure and Copula function type, until the tree structure and Copula function type of all levels are determined. The final optimized 24-dimensional vine Copula structure is as follows: Figure 3 As shown;

[0143] S32e, comparing the difference between the Cramer-von Mises statistical empirical Copula function and the constructed Copula function, selects the Copula model with the best fit:

[0144] (19)

[0145] in, To test the goodness of fit between empirical Copula functions and constructed Copula functions, and Let represent the empirical Copula distribution function and the constructed Copula distribution function, respectively.

[0146] In this embodiment, the high-dimensional hybrid vine Copula joint distribution function based on the custom structure constructed in step S3 is used to continuously derive the wind and solar power output for each hour through inverse sampling, thereby randomly generating the intraday wind and solar power output scenario. The coupling method used in this application can accurately simulate the intermittent characteristics of wind and solar power output, while preserving the complex correlation and structure between high-dimensional variables.

[0147] S4. Based on the custom structure high-dimensional hybrid vine Copula joint distribution function constructed in step S3, the wind and solar power output for each hour of the day is derived by inverse sampling, thereby randomly generating a set of wind and solar power output scenarios that retain intermittency and correlation characteristics as simulation samples.

[0148] According to one aspect of this application, step S4 further comprises:

[0149] S41, Assuming there is dimension random variable (That is, the wind and solar power output sample corresponding to 24 hours). Then the corresponding representation is a random variable. The value in for The marginal distribution function, for The joint conditional distribution function is given by joint distribution function right Taking the partial derivative, we get, which can also be simply expressed as :

[0150] (20)

[0151] Similarly, when 3-dimensional random variables are involved, for The joint conditional distribution function, obtained by taking the partial derivative of the corresponding joint distribution function with respect to the known parts, can be simply expressed as: :

[0152] (twenty one)

[0153] Mode middle The following are brief expressions of the joint conditional distribution function:

[0154] (twenty two)

[0155] in, for The joint conditional distribution function, for The joint conditional distribution function.

[0156] And so on 3D random variable, for The joint conditional distribution function can be expressed as:

[0157] (twenty three)

[0158] S42. Generating uniformly distributed data based on the Monte Carlo method. 3D random vector Iteratively solve the distribution function of the wind and solar power output sequence for each time period, let:

[0159] (twenty four)

[0160] in, For the generated random variables that follow a uniform distribution The inverse distribution function, All of these are conditional distribution functions in the joint distribution function of the high-dimensional hybrid vine Copula, which is constructed through step S32.

[0161] S43, according to formula (24) The inverse function is used to derive the power distribution function for the first time period. :

[0162] (25)

[0163] S44. From the first-layer structure of the vine Copula determined in step S32, it can be seen that the power distribution function between two adjacent time periods is connected by a certain Copula function. That is, the power distribution function of the second time period can be derived from the power distribution function of the first time period. From formula (24) Finding the inverse function yields:

[0164] (26)

[0165] The power distribution function for the second time period can be derived from the derivation principle of step S41:

[0166] (27)

[0167] (28)

[0168] Wherein, according to the definition in step S41, for joint conditional distribution function The inverse function of the concise expression of .

[0169] S45. Similarly, the power distribution function for the third time period can be derived:

[0170] (29)

[0171] (30)

[0172] Wherein, according to the definition in step S41, for joint conditional distribution function The inverse function of the concise expression of .

[0173] S46, and so on, to obtain the... Power distribution function for a given time period:

[0174] (31)

[0175] (32)

[0176] Wherein, according to the definition in step S41, for joint conditional distribution function The inverse function of the concise expression of .

[0177] S47. A new set of random numbers is obtained through the inverse sampling process of steps S42-S46. These random numbers contain some dependency information between variables. In each subsequent layer of the vine Copula, the inverse sampling process of steps S42-S46 is repeated, and the distribution function of the nodes associated with each time period is derived layer by layer based on the optimal vine Copula structure, function type, and parameters determined in step S32. Based on the variable correlation established in the previous layer, the correlation between variables is further described, finally obtaining the hourly wind and solar power output distribution function containing complete correlations. and , .in, For the first The wind power output distribution function per hour For the first Hourly wind power output variation For the first The photoelectric output power distribution function per hour For the first The hourly photovoltaic output power variation.

[0178] S48. Substitute the hourly wind and solar power output distribution functions randomly generated in step S47 into the inverse function of the hourly cumulative distribution function fitted in step S22e to obtain the hourly wind and solar power output samples (i.e., simulation samples) generated by random simulation. and , The final statistical distribution of the simulated sample and the observed sample is as follows: Figure 4 As shown in (a) and (b).

[0179] S5. Calculate the evaluation index values ​​of the observed samples and simulated samples based on the descriptive index system of wind and solar power output characteristics, and compare the evaluation index values ​​of the observed samples and the evaluation index values ​​of the simulated samples to evaluate the descriptive effect of the method on the randomness, intermittency and correlation characteristics of wind and solar power output at the hourly scale.

[0180] The feature description index system established in step S1 is used as the evaluation index. The evaluation index values ​​of the hourly-scale wind and solar power output samples (i.e., the observed samples) are calculated and compared with the evaluation index values ​​of the hourly-scale wind and solar power output samples (i.e., the simulated samples) obtained by the random simulation of the new energy power output samples (i.e., the new energy power output stochastic simulation model based on the "discrete-continuous" hybrid marginal distribution function and the high-dimensional Copula joint distribution function constructed in steps S2, S3 and S4) to evaluate the simulation effect of this method on the randomness, intermittency and correlation characteristics of hourly-scale wind and solar power output.

[0181] In the study area, by applying the random generation method for renewable energy power output based on a hybrid distribution function and a high-dimensional Copula function proposed in this invention, hourly wind and solar power output scenario sets can be quickly simulated and generated. The hourly wind and solar power output values ​​are accurately simulated while retaining their intermittent and correlation characteristics, thereby improving the operating efficiency of the new power system with renewable energy as the main component.

[0182] In this embodiment, the hourly wind and solar power output can be obtained through all the above steps, providing effective and reliable reference information for the operation of a new type of power system based on new energy sources, and has high technical innovation and practicality. The preferred embodiments of the present invention have been described in detail above; however, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and all such equivalent transformations fall within the protection scope of the present invention.

Claims

1. A method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function, characterized in that, Includes the following steps: S1. Based on the hourly wind and solar natural resource data and engineering data of the new energy base, calculate the hourly wind and solar power output samples of the base as observation samples, and establish a descriptive index system for wind and solar power output characteristics. S2. Construct a hybrid marginal distribution function based on the Dirac and Kernal functions, divide the daily hourly wind and solar power output observation samples into discrete and continuous parts, and fit the power output distribution of the discrete and continuous parts based on the Dirac and Kernal functions in the hybrid marginal distribution function. S3. A high-dimensional hybrid vine Copula is used to connect the hybrid marginal distribution function. The corresponding vine structure, Copula function type and parameters are selected. The goodness-of-fit evaluation index is used to evaluate the fitting effect of the distribution function, thereby constructing a custom structure high-dimensional hybrid vine Copula joint distribution function that describes the correlation characteristics between the output power of each hour in the day. S4. Based on the constructed custom structure high-dimensional hybrid vine Copula joint distribution function, the wind and solar power output for each hour of the day is derived by inverse sampling, thereby randomly generating a set of wind and solar power output scenarios that retain intermittency and correlation characteristics as simulation samples. S5. Calculate the evaluation index values ​​of the observed samples and simulated samples based on the descriptive index system of wind and solar power output characteristics, and compare the evaluation index values ​​of the observed samples and the evaluation index values ​​of the simulated samples to evaluate the descriptive effect of the method on the randomness, intermittency and correlation characteristics of wind and solar power output at the hourly scale.

2. The method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function as described in claim 1, characterized in that, Step S1 includes the following steps: S11. Collect hourly wind and solar natural resource data and engineering data of the new energy base. The wind and solar natural resource data includes hourly wind speed, air density, altitude, solar radiation and temperature. The engineering data includes the number of wind turbines, installed capacity, output coefficient, cut-in wind speed, rated wind speed, cut-out wind speed, hub area, hub height, wind speed conversion coefficient, number of photovoltaic panels, installed capacity, area and heat loss efficiency coefficient. S12. Calculate the hourly wind and solar power output of the new energy base based on the calculation model of a single wind turbine and photovoltaic power output, wind and solar natural resource data and engineering data; S13. Establish a descriptive index system for the randomness, intermittency, and correlation characteristics of wind and solar power output at the hourly scale; including: (1) The randomness characteristics of wind and solar power output on an hourly scale are measured by the sample mean, standard deviation, coefficient of variation, kurtosis coefficient and power output volatility. (2) The power output activity is used to measure the intermittent characteristics of wind and solar power output on an hourly scale; (3) Kendall correlation coefficient is used to measure the temporal correlation characteristics between wind and solar power output on an hourly scale.

3. The method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function as described in claim 1, characterized in that, Step S2 specifically includes: S21. Compile the hourly output power of wind power and photovoltaic power into a sample group with a length of [missing information]. Dimensions The sample spaces are respectively of 3D random variables ; This refers to the installed capacity of wind power, i.e., the maximum output value of wind power. This refers to the installed capacity of the optoelectronic system, i.e., the maximum output value of the optoelectronic system. , These represent the observation samples consisting of wind power and solar power output for each hour of the day. S22. Considering the upper and lower boundary properties of wind and solar power output at the hourly scale, the Dirac function and the Kernal function are introduced to represent the upper and lower boundary values ​​and continuous values ​​in wind and solar power output, respectively. A hybrid marginal distribution function is constructed to describe the double boundary distribution characteristics of wind and solar power output variables at the hourly scale. S23. Use the root mean square error (RMSE) and goodness-of-fit R². 2 The evaluation criteria assess the fitting effect of the hybrid marginal distribution function constructed in step S22 on the distribution of the dual-boundary wind and solar power output variables at the hourly scale, which includes a large number of zero values ​​and rated output values.

4. The method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function as described in claim 3, characterized in that, Step S22 specifically includes: (1) Construct a mixed probability density function for the hour-scale double-boundary wind and solar power output variables, expressed as: ； ； in, Let be the probability density function of wind power output; Let be the probability density function of the photoelectric output power; express The Dirac function at that location, and They represent and The Dirac function at the location; , These represent the Kernal functions obtained by fitting the power output variables of wind power and solar power, respectively. For random variables The value in the random variable For a certain period of time The output power of wind power or solar power; This is the intermittency coefficient; (2) The intermittency coefficient is determined by the proportion of the number of days with zero wind or solar power output and the number of days with rated power in a certain period to the total number of days in that period; the expression is: ； (3) The constructed probability density function of wind power and photovoltaic power output hybrid type The final cumulative distribution function, i.e., the mixed marginal distribution function, is derived as follows: , : ； ； in, These are the cumulative power distribution functions for wind power and solar power output, respectively. The definition and derivation expression of the random variable is given. Less than or equal to The probability of; The derivation expressions for the cumulative distribution functions when the power output of wind power and solar power is greater than 0 are respectively given. random variables Less than or equal to The probability of.

5. The method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function according to claim 4, characterized in that, Dirac function The probability density used to represent the discrete part is expressed as: ； ； Kernal function The probability density used to represent the continuous portion is expressed as: ； ； in, For a certain period of time Number of wind power or solar power output samples; Indicates the first Individual wind power / solar power output values; For kernel function parameters, The optimal parameters; For random variables The standard deviation.

6. The method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function according to claim 1, characterized in that, Step S3 specifically includes: (1) Treat the hourly wind and solar power output variables as nodes and calculate the Kendall correlation coefficients between all nodes in the first-level tree structure and the remaining nodes in turn. ; (2) Estimate the Copula function type and parameter value connecting two nodes in the first-level tree structure, and use the AIC criterion to determine the optimal Copula function type from the aspects of goodness of fit and complexity; ,in, and These represent the number of parameters of the currently selected Copula function and the value of the maximum likelihood function of the Copula function, respectively. (3) The parameters of the Copula function are determined by the maximum likelihood estimation method; (4) After determining the first-level Copula function, sample the Copula function of each side in turn to obtain the corresponding node value of the next level tree. Repeat the above steps (1) to (3) to determine the tree structure and Copula function type of the next level until the tree structure and Copula function type of all levels are determined. (5) Based on the difference between the Cramer-von Mises statistical empirical Copula function and the constructed Copula function, the Copula model with the best fitting effect is selected as: ,in, To test the goodness of fit between empirical Copula functions and constructed Copula functions, and Let represent the empirical Copula distribution function and the constructed Copula distribution function, respectively.

7. The method for stochastically generating new energy power output based on a hybrid distribution function and a high-dimensional Copula function as described in claim 1, characterized in that, Step S4 specifically includes: (1) Derive hourly scale wind and solar power output samples The corresponding marginal distribution function, joint distribution function, and joint conditional distribution function; (2) Generating uniformly distributed data based on the Monte Carlo method 3D random vector The distribution function of the wind and solar power output sequence for each time period is solved iteratively. ； in, Corresponding representation The value in; For the generated random variables that follow a uniform distribution The inverse distribution function, All of these are the conditional distribution functions in the joint distribution function of the high-dimensional hybrid vine Copula, which is a custom structure constructed by the user. (3) Based on the formula in step (2), derive the power distribution function for each time period by finding the inverse function; then the first... Power distribution function during time period for; ； in, for joint conditional distribution function The inverse function of the concise expression; (4) Repeat the sampling process of steps (2) to (3) in each layer of the subsequent vine Copula structure, and derive the distribution function of the nodes associated with each time period layer by layer according to the optimal vine Copula structure, function type and parameters, and finally obtain the distribution function of wind and solar power output for each hour containing complete correlation. (5) By substituting the randomly generated hourly wind and solar power output distribution function into the inverse function of the fitted hourly cumulative wind and solar power output distribution function, the hourly wind and solar power output samples generated by random simulation are obtained, i.e., the simulation samples.

8. A random power generation system for new energy sources based on a hybrid distribution function and a high-dimensional Copula function, characterized in that, include: The data and indicator system unit is used to calculate the wind and solar power output samples at the hourly scale of the new energy base based on the wind and solar natural resource data and engineering data of the base, and to establish a descriptive indicator system for the characteristics of wind and solar power output. The hybrid marginal distribution function construction unit is used to construct a hybrid marginal distribution function based on the Dirac function and the Kernal function. It divides the wind and solar power output observation samples of each hour of the day into discrete and continuous parts, and fits the power output distribution of the discrete and continuous parts based on the Dirac function and the Kernal function in the hybrid marginal distribution function. The joint distribution function construction unit is used to connect the hybrid marginal distribution function with a high-dimensional hybrid vine Copula. The corresponding vine structure, Copula function type and parameters are selected, and the goodness-of-fit evaluation index is used to evaluate the fitting effect of the distribution function, thereby constructing a custom structure high-dimensional hybrid vine Copula joint distribution function that describes the correlation characteristics between the output power of each hour in the day. The wind and solar power output simulation unit is used to derive the wind and solar power output for each hour of the day by inverse sampling based on the constructed custom structure high-dimensional hybrid vine Copula joint distribution function, thereby randomly generating a set of wind and solar power output scenarios that retain intermittency and correlation characteristics as simulation samples. The evaluation unit is used to calculate the evaluation index values ​​of the observed samples and the simulated samples based on the descriptive index system of wind and solar power output characteristics, and compare the evaluation index values ​​of the observed samples and the evaluation index values ​​of the simulated samples to evaluate the descriptive effect of the method on the randomness, intermittency and correlation characteristics of wind and solar power output at the hourly scale.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor. When the computer program / instruction is executed by the processor, it implements the steps of the method for random generation of new energy power output based on a hybrid distribution function and a high-dimensional Copula function as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked, are used to execute the steps of the method for random generation of new energy power output based on a hybrid distribution function and a high-dimensional Copula function as described in any one of claims 1-7.

Images