Statistical modeling method and device for evaluating demand of high-proportion new energy on frequency modulation capacity of power system, and medium
By collecting, preprocessing, and modeling the Laplace distribution of wind and solar power data, the problem of insufficient frequency regulation capacity reservation in high-proportion renewable energy power systems has been solved, enabling accurate assessment and fine-grained analysis of frequency regulation demand.
Patent Information
- Application Number
- CN202511172207.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-21
- Publication Date
- 2025-11-07
AI Technical Summary
The fluctuating characteristics of wind and solar power in high-proportion renewable energy power systems lead to insufficient and wasted traditional frequency regulation capacity, and there is a lack of effective evaluation methods.
By collecting and preprocessing wind power and photovoltaic data, the fluctuation components are extracted, and statistical modeling is performed using power per-unitization and Laplace distribution to calculate the frequency regulation capacity demand.
It enables accurate assessment of frequency regulation capacity demand in high-proportion renewable energy power systems, adapts to changes in grid wind and photovoltaic installed capacity, and provides fine-grained annual and daily frequency regulation demand analysis.
Smart Images
Figure CN120914831A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of new energy power system, and particularly relates to a statistical modeling method for evaluating the frequency modulation capacity demand of a high-proportion new energy power system, a device and a medium. BACKGROUND
[0002] Power system frequency modulation is mainly used to cope with short-term power generation and load fluctuations in the power grid, and to suppress power grid frequency fluctuations, which is an important part of power grid operation control. With the large-scale access of new energy, the proportion of wind power and photovoltaic power, as representatives of new energy, in China's power system continues to increase. New energy has strong volatility much higher than load, which makes it difficult for the traditional power grid dispatching department to reserve frequency modulation capacity according to load fluctuations to meet operational requirements.
[0003] High-proportion new energy power systems often contain a large number of wind power and photovoltaic units at the same time. The wind power and photovoltaic power generation and fluctuation characteristics not only change dramatically with the season and weather, but also differ greatly in the shape of the daily power generation curve and the peak and valley periods. There is still a lack of mature evaluation methods to model and analyze the frequency modulation capacity demand brought by new energy fluctuations and to reserve them in the scheduling and operation mode arrangement. SUMMARY
[0004] In view of the above problems, the present application is proposed.
[0005] Therefore, the technical problem solved by the present application is: how to accurately evaluate the fluctuation characteristics of wind power and photovoltaic power in a high-proportion new energy power system through a statistical modeling method to scientifically determine the frequency modulation capacity demand and solve the problems of insufficient frequency modulation capacity reservation and waste caused by the strong randomness of traditional methods.
[0006] To solve the above technical problems, the present application provides the following technical solutions: a statistical modeling method for evaluating the frequency modulation capacity demand of a high-proportion new energy power system, comprising,
[0007] Collecting wind power data and photovoltaic data and preprocessing them; extracting fluctuation components from the preprocessed wind power data and photovoltaic data to obtain fluctuation sequence data; and statistically modeling the fluctuation sequence data through power normalization and Laplace distribution to obtain a wind power and photovoltaic output fluctuation statistical model.
[0008] As a preferred solution of the statistical modeling method for evaluating the frequency modulation capacity demand of a high-proportion new energy power system, the preprocessing includes reverse compensation of missing data caused by external factors by superimposing the missing amount to restore the theoretical available power value.
[0009] Discrete abnormal data is corrected by mean filling.
[0010] Delete and mark the data records with continuous missing.
[0011] Divide all wind power and photovoltaic power records by the corresponding time period of the grid wind power and photovoltaic installed capacity to obtain the normalized power generation power time series.
[0012] For discrete abnormal data, mean filling is used for correction.
[0013] Delete and mark the data records with continuous missing.
[0014] Divide all wind power and photovoltaic power records by the corresponding time period of the grid wind power and photovoltaic installed capacity to obtain the normalized power generation power time series.
[0015] As a preferred scheme of the statistical modeling method for evaluating the demand of high-proportion new energy on the frequency modulation capacity of the power system according to the application, wherein the fluctuation component extraction comprises,
[0016] Select the fluctuation amount statistical interval.
[0017] Point-by-point calculation of wind power and photovoltaic total power generation power fluctuation amount samples.
[0018] As a preferred scheme of the statistical modeling method for evaluating the demand of high-proportion new energy on the frequency modulation capacity of the power system according to the application, wherein the power normalization comprises,
[0019] Divide all samples by time.
[0020] The statistical modeling of the fluctuation sequence data comprises,
[0021] The probability density function and the cumulative distribution function corresponding to each sample subset data probability density are obtained by fitting the Laplace distribution.
[0022] As a preferred scheme of the statistical modeling method for evaluating the demand of high-proportion new energy on the frequency modulation capacity of the power system according to the application, wherein the statistical modeling of the fluctuation sequence data comprises,
[0023] The probability density function and the cumulative distribution function corresponding to each sample subset data probability density are obtained by fitting the Laplace distribution.
[0024] As a preferred scheme of the statistical modeling method for evaluating the demand of high-proportion new energy on the frequency modulation capacity of the power system according to the application, wherein the data probability density is expressed as,
[0025]
[0026] Wherein, μ is the location parameter, and λ represents the scale parameter.
[0027] As a preferred scheme of the statistical modeling method for evaluating the high-proportion new energy power system frequency modulation capacity demand, wherein: the wind power and photovoltaic output fluctuation amount statistical model is inputted with the predicted date and the total installed capacity C1 of wind power and the total installed capacity C2 of photovoltaic on the day.
[0028] The wind power fluctuation probability density function formed in the last section is f W,m,h (x, λ), and the photovoltaic fluctuation probability density function is f S,m,h (x, λ).
[0029] The wind power fluctuation probability model set {f W,M,h (x, λ)|h=0, 1,..., 23} and the photovoltaic fluctuation probability model set {f S,M,h (x, λ)|h=0, 1,..., 23} corresponding to the 24 hours of the month are taken.
[0030] The wind power and photovoltaic total power probability density function set {f WS，M，h (x, λ)|h=0, 1,..., 23} of the predicted day is calculated, and the calculation formula is as follows:
[0031]
[0032] Wherein, is a convolution symbol.
[0033] According to the obtained 24-hour new energy fluctuation amount probability density function f WS，M，h (x, λ), the cumulative distribution function F WS,M,h (x, λ) is calculated by integrating each function.
[0034] The new energy power fluctuation amplitude ΔP 1-α is calculated under the (1-α) confidence interval, which is the frequency modulation capacity demand of each period, and the new energy power fluctuation amplitude ΔP 95％ = max{|F WS,M,h -1 (x, λ)|h=0, 1,..., 23}, forming the predicted result of the frequency modulation capacity sequence.
[0035] The application provides a computer device, comprising a memory and a processor, the memory stores a computer program, characterized in that the processor implements the steps of the statistical modeling method for evaluating the high-proportion new energy power system frequency modulation capacity demand when executing the computer program.
[0036] The application provides a computer readable storage medium, which stores a computer program, and the computer program is characterized in that the computer program is executed by a processor to realize the steps of the statistical modeling method for evaluating frequency modulation capacity demand of a high proportion of new energy on a power system.
[0037] The application has the following beneficial effects: the application provides a data-driven frequency modulation demand evaluation method for a high proportion of new energy.
[0038] The modeling and analysis method has the following advantages: (1) the modeling granularity is fine, and the frequency modulation demand of 12 months in a year and 24 hours in a day can be independently modeled and evaluated. BRIEF DESCRIPTION OF DRAWINGS
[0039] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description.
[0040] Figure 1 The application provides a statistical modeling method for evaluating frequency modulation capacity demand of a high proportion of new energy on a power system.
[0041] Figure 2 The application provides a statistical modeling method for evaluating frequency modulation capacity demand of a high proportion of new energy on a power system.
[0042] Figure 3 The application provides a statistical modeling method for evaluating frequency modulation capacity demand of a high proportion of new energy on a power system.
[0043] Figure 4 The application provides a statistical modeling method for evaluating frequency modulation capacity demand of a high proportion of new energy on a power system. DETAILED DESCRIPTION
[0044] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.
[0045] Embodiment 1, reference Figure 1 For an embodiment of the present application, the embodiment provides a statistical modeling method for evaluating the frequency modulation capacity demand of a high proportion of new energy to a power system, comprising:
[0046] Further, S1: collecting wind power data and photovoltaic data and preprocessing.
[0047] In the embodiment of the present application, the preprocessing includes, for the known wind curtailment and light curtailment time period caused by peak shaving and channel obstruction, adding the power generation power record of the corresponding time period to the wind curtailment and light curtailment record of each hour to restore the available power value.
[0048] The discrete abnormal data is corrected by mean filling.
[0049] The continuous missing data record is deleted and marked.
[0050] In the embodiment of the application, all wind power and photovoltaic power records are divided by the corresponding time period of the grid wind power and photovoltaic installed capacity to obtain the normalized power generation power time sequence, denoted as PW t , PS t , respectively corresponding to the grid wind power and photovoltaic power sequence.
[0051] Further, S2: extracting fluctuation component from the preprocessed wind power data and photovoltaic data to obtain fluctuation sequence data.
[0052] In the embodiment of the present application, the fluctuation component extraction includes.
[0053] The fluctuation amount statistical interval D is selected, and D=5 minutes is recommended according to the typical frequency modulation coverage time scale.
[0054] The total wind power and photovoltaic power fluctuation sample is calculated point by point: ΔPW t-D = PW t -PW t-D , ΔPS t-D = PS t -PS t-D .
[0055] Wherein, ΔPW t-D represents the total wind power fluctuation sample, and ΔPSt-D representing the sample of the total photovoltaic power fluctuation.
[0056] Further, S3: statistical modeling of the fluctuation sequence data by power normalization and Laplace distribution to obtain the wind power and photovoltaic power fluctuation statistical model.
[0057] In the embodiments of the present application, the power normalization includes,
[0058] All samples are divided into 12*24 subsets according to time, wherein the mth subset contains the fluctuation sample of the mth month from h to h+1 o'clock time interval of any year.
[0059] Statistical modeling of the fluctuation sequence data includes,
[0060] The probability density of each sample subset data is fitted by the Laplace distribution to obtain the corresponding probability density function and cumulative distribution function.
[0061] The data probability density is represented as,
[0062]
[0063] Wherein, μ is the location parameter, and λ represents the scale parameter.
[0064] Since the location parameter μ=0 in the new energy fluctuation statistical model, only the scale parameter λ of the Laplace distribution needs to be estimated, and f(x, λ) is used. It should be noted that since there is no light at night within a day, the fluctuation of photovoltaic power in the corresponding time period is directly taken as zero, and there is no need to establish a statistical model.
[0065] According to the established wind power and photovoltaic power fluctuation statistical model. Since the distribution of the sum of two random variables and is not equal to the sum of the distributions of the variables, the calculation of the new energy total power fluctuation distribution function needs to convolve the wind power fluctuation probability density function and the photovoltaic power fluctuation probability density function. In addition, when applying the model in the previous section to predict the frequency modulation capacity demand, the development and change of the wind power and photovoltaic installed capacity also need to be considered, and the calculation is performed according to the installed capacity of the prediction day. The specific steps are as follows:
[0066] Based on the wind power and photovoltaic power fluctuation statistical model, the prediction date and the total installed capacity of the wind power C1 and the total installed capacity of the photovoltaic C2 on the day are input, and the model of the wind power fluctuation probability density function formed in the previous section is f W,m,h (x, λ) a total of 288, and the photovoltaic fluctuation probability density function is f S,m,h (x, λ) a total of 146, and the non-illumination period is considered as 0.
[0067] The wind power fluctuation probability model set {f W,M,h (x, λ) | h = 0, 1,..., 23} and the photovoltaic fluctuation probability model set {f S,M,h (x, λ) | h = 0, 1,..., 23} are taken.
[0068] The probability density function set {f WS，M，h (x, λ) | h = 0, 1,..., 23} of the total wind power and photovoltaic power on the prediction day is calculated according to the following formula:
[0069]
[0070] wherein, is a convolution symbol.
[0071] According to the obtained 24-hour new energy fluctuation probability density function f WS,M,h (x, λ), the cumulative distribution function F WS,M,h (x, λ) is calculated by integrating each function.
[0072] The new energy power fluctuation amplitude ΔP 1-α in the (1-α) confidence interval is calculated, which is the frequency modulation capacity demand of each period, and the new energy power fluctuation amplitude ΔP 95％ = max{|F WS,M,h -1 (α / 2) |}, forming the prediction result of the frequency modulation capacity sequence.
[0073] Embodiment 2, referring to Figures 2-4 , provides a statistical modeling method for evaluating the frequency modulation capacity demand of a high proportion of new energy to an electric power system, and scientific demonstration is carried out through experiments to verify the beneficial effects of the present application.
[0074] This section is based on the statistical analysis of short-term fluctuation characteristics of the annual power records of wind power and photovoltaic power in a certain province. The data records are 1 minute per point. Taking D = 5 minutes, the normalized power time series is calculated, and the wind power / photovoltaic power fluctuation sample ΔPW t-D , ΔPS t-D .
[0075] Given PW t , PS t , next, the relationship between the fluctuation distribution and the month and time is further analyzed, and the steps are as follows:
[0076] Firstly, the annual fluctuation sample is divided into 12*24 = 288 sample subsets, wherein the (m, h) sample group contains all the fluctuation samples from h to h+1 points in the mth month.
[0077] Next, Laplace distribution fitting was performed on each wind power sample subset using the method described above. This process was repeated for all 288 sample subsets to obtain f. W,M,h (x,λ), the variance 2λ is obtained. 2 As shown in Table 6-1 below.
[0078] Table 6-1 Statistical Table of Variance of 5-Minute Laplace Distribution of Wind Power in a Certain Province Throughout the Year
[0079]
[0080]
[0081] Taking the period from 3:00 to 4:00 in November as an example, the fitting of the sample frequency plot and the Laplace distribution function are as follows: Figure 2 As shown in the left figure, its fluctuation probability density function is f. W,11,3 The probability density function has a value of (x, 0.0118), a variance of 0.0003, and a fitting error (average absolute relative error) of 0.7139. The right figure shows the cumulative probability distribution corresponding to this probability density function.
[0082] Similarly, the photovoltaic fluctuation sample set was divided into 288 subsets. After removing periods without sunlight, 146 subsets were retained. Each subset was fitted with a Laplace distribution, and the above method was applied to each of these 146 subsets to obtain f. S,M,h (x,λ), the variance 2λ is obtained. 2 As shown in Table 6-2 below:
[0083] Table 6-2 Statistical Table of Variance of 5-Minute Laplace Distribution of Photovoltaic Power Stations in a Certain Province Throughout the Year
[0084]
[0085]
[0086] by Figure 3 For example, to fit the Laplace distribution of sample data from 11:00 to 12:00 in September, the fluctuation probability density function is obtained as f. S,9,11 (x, 0.0922), with a variance of 0.0170 and a fitting error of 0.1189.
[0087] Using the above model, we estimate the frequency regulation capacity demand of a planned power grid in January.
[0088] Given that the installed capacities of wind power and photovoltaic power in the planned power grid are C1 = 50 and C2 = 50 (MW), respectively, we take the Laplace distribution parameters of the 24 wind power fluctuations in the first column of Table 6-1 and the 24 photovoltaic power fluctuations in the first column of Table 6-2, substitute them pairwise into formula (2), and convolve them to obtain the Laplace distribution parameters of the total power fluctuation of new energy sources in the planned power grid for the 24 time periods throughout January. 2 The results are shown in Table 6-3.
[0089] Table 6-3 Statistical Table of Variance of 5-Minute Laplace Distribution of New Energy in a Certain Region Throughout the Year
[0090] o'clock 0 1 2 3 4 5 6 7 2λ 2 ]]> 1.8751 2.4049 3.8909 2.1071 1.8170 1.9083 3.3891 1.3871 o'clock 8 9 10 11 12 13 14 15 2λ 2 ]]> 1.5041 3.9190 6.5270 8.6653 9.4481 11.1089 11.8435 9.7891 o'clock 16 17 18 19 20 21 22 23 2λ 2 ]]> 2.4396 2.2431 1.9384 1.8296 1.5191 1.5803 2.3164 1.8093
[0091] The probability density distribution of new energy fluctuations at points 9-10 is used as an example for illustration. According to Table 6-3, 2λ 2 =3.919, therefore λ = 1.3998. f WS,1,9 (x, 1.3998) and cumulative probability distribution plot F WS,1,9 (x, 1.3998) Figure 4 As shown, the obtained ΔP 95％ The value is 4.2831 (MW), which represents the demand for new energy frequency regulation capacity during this period of the month.
[0092] Following the steps above, the frequency regulation capacity demand sequence for this power grid from 0:00 to 23:00 in January is calculated as follows:
[0093] {2.9671 3.0512 3.1084 2.1724 2.0676 2.3133 2.9546 2.3490 2.8149 4.2831 5.1269 5.4807 5.5283 5.9493 5.9023 5.1602 3.3500 3.2283
[0095] 2.8845 2.6567 2.4870 2.5399 2.9145 2.8319}, unit (MW).
[0096] This embodiment also provides an electronic device applicable to a statistical modeling method for assessing the frequency regulation capacity demand of a power system from a high proportion of renewable energy sources, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the statistical modeling method for assessing the frequency regulation capacity demand of a power system from a high proportion of renewable energy sources as proposed in the above embodiment.
[0097] The embodiment also provides a storage medium, which stores a computer program, and the computer program is executed by a processor to implement a statistical modeling method for evaluating frequency modulation capacity demand of a power system under high proportion of new energy as proposed in the above embodiment.
[0098] The storage medium proposed in the embodiment belongs to the same inventive concept as the statistical modeling method for evaluating frequency modulation capacity demand of a power system under high proportion of new energy proposed in the above embodiment, and the technical details not described in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.
[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary universal hardware, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH memory, a hard disk, or an optical disc, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.
[0100] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.
Claims
1. A statistical modeling method for evaluating the demand of high proportion of new energy on the frequency modulation capacity of a power system, characterized in that: The application relates to a statistical modeling method for evaluating the demand of a high-proportion new energy on a power system frequency modulation capacity. Wind power data and photovoltaic data are collected and pretreated; Fluctuation component extraction is performed on the pretreated wind power data and photovoltaic data to obtain fluctuation sequence data; Statistical modeling is performed on the fluctuation sequence data through power normalization and Laplace distribution to obtain a wind power and photovoltaic output fluctuation statistical model.
2. The statistical modeling method for assessing the demand of high proportion of new energy on power system frequency modulation capacity according to claim 1, characterized in that: The pretreatment includes reverse compensation of missing data caused by external factors, and recovery of theoretical available power values through superposition of missing data records; Mean filling is adopted to correct discrete abnormal data; Continuous missing data records are deleted and marked; All wind power and photovoltaic power records are divided by the corresponding time period of the grid wind power and photovoltaic installed capacity to obtain normalized power generation time series.
3. The statistical modeling method for assessing the demand of high proportion of new energy on power system frequency modulation capacity according to claim 2, characterized in that: The fluctuation component extraction includes, Selection of fluctuation statistical intervals; Point-by-point calculation of wind power and photovoltaic total power generation fluctuation samples.
4. The statistical modeling method for assessing the demand of high proportion of new energy on power system frequency modulation capacity according to claim 3, characterized in that: The power normalization includes, Division of all samples according to time; The statistical modeling of the fluctuation sequence data includes, Fitting of the probability density of each sample subset data through Laplace distribution to obtain corresponding probability density functions and cumulative distribution functions.
5. The statistical modeling method for assessing the demand of frequency modulation capacity of power system with high proportion of new energy according to claim 4, characterized in that: The data probability density is expressed as, Wherein, mu is a location parameter, and lambda represents a scale parameter.
6. The statistical modeling method for assessing the demand of high proportion of new energy on power system frequency modulation capacity according to claim 5, characterized in that: The wind power and photovoltaic output fluctuation statistical model is input with a prediction date and total wind power installed capacity C1 and total photovoltaic installed capacity C2 of the day; The wind power fluctuation probability density function formed in the last section is f W,m,h (x, λ), and the photovoltaic fluctuation probability density function is f S,m,h (x, λ); Record the wind power fluctuation probability model set {f W,M,h (x,λ)|h=0,1,...,23} and the photovoltaic fluctuation probability model set {f S,M,h (x,λ)|h=0,1,...,23}. A set of probability density functions {f WS,M,h (x, λ) | h = 0, 1,..., 23}, the calculation formula is as follows: wherein is the convolution symbol; According to the obtained 24-hour new energy fluctuation probability density function f WS,M,h (x, λ), the cumulative distribution function F WS,M,h (x, λ) is calculated by integrating each function. The new energy power generation power fluctuation amplitude ΔP under the calculation (1-α) confidence interval 1-α As the frequency modulation capacity demand of each period, the new energy power generation power fluctuation amplitude The prediction result of forming the frequency modulation capacity sequence. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the statistical modeling method for evaluating the demand of a high-proportion new energy on a power system frequency modulation capacity.
8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the statistical modeling method for evaluating the demand of a high-proportion new energy on a power system frequency modulation capacity.