Photovoltaic typical output curve extraction method, device, equipment and medium

Abstract

The present disclosure relates to a method, device, equipment and medium for extracting a typical photovoltaic output curve. The method for extracting a typical photovoltaic output curve includes: obtaining the original historical output data of the photovoltaic power station, and performing data preprocessing on the original historical output data to obtain a historical daily output curve; performing a first category classification process on the historical daily output curve based on a mean clustering algorithm of a first distance indicator to obtain a corresponding first weather category, wherein the first weather category includes a high fluctuation category; performing a second category classification process on the target historical daily output curve under the high fluctuation category based on a mean clustering algorithm of a second distance indicator to obtain a corresponding second weather category; respectively extracting the photovoltaic typical output curves corresponding to the first weather category and the second weather category. According to the embodiments of the present disclosure, the photovoltaic typical output curves corresponding to different weather categories can be accurately extracted.

Description

Technical Field

[0001] The present disclosure relates to the application field of new energy output prediction technology, and in particular to a method, device, equipment and medium for extracting a typical photovoltaic output curve. Background Art

[0002] Photovoltaic power generation output is affected by many factors such as weather conditions, irradiance, and ambient temperature, and has a high degree of uncertainty.

[0003] The traditional single-stage clustering method based on the Euclidean distance indicator can only distinguish the mean value of the photovoltaic output curve, but cannot distinguish the time-sharing fluctuations and overall changes of the photovoltaic output curve under different weather scenarios, making it difficult to achieve refined extraction of the typical photovoltaic output curve. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method, device, equipment and medium for extracting a typical photovoltaic output curve.

[0005] In a first aspect, the present disclosure provides a method for extracting a typical photovoltaic output curve, comprising:

[0006] Acquire original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain a historical daily output curve;

[0007] Performing a first category classification process on the historical daily output curve based on a mean value clustering algorithm of a first distance index to obtain a corresponding first weather category, wherein the first weather category includes a high fluctuation category;

[0008] Based on the mean value clustering algorithm of the second distance index, the target historical daily output curve under the high fluctuation category is divided into the second category to obtain the corresponding second weather category;

[0009] The photovoltaic typical output curves corresponding to the first weather category and the second weather category are extracted respectively.

[0010] In a second aspect, the present disclosure provides a photovoltaic typical output curve extraction device, comprising:

[0011] The first processing module is used to obtain the original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain a historical daily output curve;

[0012] A second processing module is used to perform a first category classification process on the historical daily output curve based on a mean value clustering algorithm of a first distance indicator to obtain a corresponding first weather category, wherein the first weather category includes a high fluctuation category;

[0013] A third processing module is used to perform a second classification process on the target historical daily output curve under the high fluctuation category based on a mean value clustering algorithm of a second distance indicator to obtain a corresponding second weather category;

[0014] The curve extraction module is used to extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category respectively.

[0015] In a third aspect, the present disclosure provides a photovoltaic typical output curve extraction device, comprising:

[0016] processor;

[0017] A memory for storing executable instructions;

[0018] The processor is used to read executable instructions from the memory and execute the executable instructions to implement the photovoltaic typical output curve extraction method of the first aspect.

[0019] In a fourth aspect, the present disclosure provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor implements the photovoltaic typical output curve extraction method of the first aspect.

[0020] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0021] The photovoltaic typical output curve extraction method, device, equipment and medium of the disclosed embodiments can obtain the original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain the historical daily output curve, and then perform a first category classification process on the historical daily output curve based on the mean clustering algorithm of the first distance index to obtain the corresponding first weather category, and the first weather category includes a high fluctuation category, and then perform a second category classification process on the target historical daily output curve under the high fluctuation category based on the mean clustering algorithm of the second distance index to obtain the corresponding second weather category, and finally extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category respectively. Thus, the historical daily output curve is subjected to a two-stage clustering process to extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category, so as to accurately extract the photovoltaic typical output curves corresponding to different weather categories. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0023] Figure 1 A schematic diagram of a process for extracting a typical photovoltaic output curve provided by an embodiment of the present disclosure;

[0024] Figure 2 A schematic diagram of the structure of a photovoltaic typical output curve extraction device provided in an embodiment of the present disclosure;

[0025] Figure 3 A schematic diagram of the structure of a photovoltaic typical output curve extraction device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0027] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0028] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0029] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0030] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0031] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0032] In order to solve the above problems, the embodiments of the present disclosure provide a method, device, equipment and medium for extracting a typical photovoltaic output curve. Figure 1 The method for extracting a typical photovoltaic output curve provided in an embodiment of the present disclosure is described in detail.

[0033] Figure 1 A schematic flow chart of a method for extracting a typical photovoltaic output curve provided by an embodiment of the present disclosure is shown.

[0034] In the embodiment of the present disclosure, the photovoltaic typical output curve extraction method can be executed by an electronic device, wherein the electronic device can include but is not limited to devices such as computer devices, cloud servers or cloud server clusters.

[0035] like Figure 1 As shown, the method for extracting a typical photovoltaic output curve may include the following steps.

[0036] S110 , obtaining original historical output data of the photovoltaic power station, and performing data preprocessing on the original historical output data to obtain a historical daily output curve.

[0037] In the disclosed embodiment, the electronic device may obtain original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain a historical daily output curve.

[0038] Optionally, the original historical output data may be daily time-series output data collected from the photovoltaic power station within a certain period of time.

[0039] Optionally, the historical daily power output curve may be a curve showing that the output power of the photovoltaic power station changes with time in different time periods.

[0040] Specifically, the electronic device can obtain the original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data, and obtain the historical daily output curve through the processed data.

[0041] S120. Perform a first category classification process on the historical daily output curve based on a mean clustering algorithm of a first distance index to obtain a corresponding first weather category, where the first weather category includes a high fluctuation category.

[0042] In an embodiment of the present disclosure, the electronic device may perform a first category classification process on the historical daily output curve based on a mean clustering algorithm of a first distance indicator to obtain a corresponding first weather category.

[0043] Optionally, the mean clustering algorithm of the first distance index may be a fuzzy C-means clustering algorithm based on delta distance.

[0044] Optionally, the first weather category may include a high fluctuation category and a non-high fluctuation category. For example, the non-high fluctuation category may be "sunny", "cloudy", "few clouds", etc., which are not limited here.

[0045] Specifically, the electronic device can perform a first category classification process on the historical daily output curve based on a mean clustering algorithm of a first distance indicator, such as a fuzzy C-means clustering algorithm based on δ distance, to obtain a corresponding first weather category. The first weather category can include non-high fluctuation categories (such as "sunny", "cloudy", "few clouds") and high fluctuation categories.

[0046] S130. Perform a second classification process on the target historical daily output curve under the high fluctuation category using a mean value clustering algorithm based on a second distance index to obtain a corresponding second weather category.

[0047] In an embodiment of the present disclosure, the electronic device may perform a second category classification process on the target historical daily output curve under the high fluctuation category based on a mean clustering algorithm of a second distance indicator to obtain a corresponding second weather category.

[0048] Optionally, the mean clustering algorithm of the second distance index may be a fuzzy C-means clustering algorithm based on β distance.

[0049] Optionally, the second weather category may be “cloudy to sunny”, “cloudy”, “sunny to cloudy”, “cloudy to sunny”, “sunny to cloudy”, etc., which is not limited here.

[0050] Specifically, the electronic device can perform a second category division processing on the target historical daily output curve under the high fluctuation category based on the mean clustering algorithm of the second distance indicator, that is, the fuzzy C-means clustering algorithm based on the β distance, to obtain the corresponding second weather category, such as "cloudy to clear", "cloudy", "sunny to cloudy", "cloudy to clear", "sunny to cloudy", etc.

[0051] S140. Extract photovoltaic typical output curves corresponding to the first weather category and the second weather category respectively.

[0052] In the embodiment of the present disclosure, the electronic device may respectively extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category.

[0053] Optionally, the photovoltaic typical output curve reflects the characteristics of photovoltaic power generation output changing over time.

[0054] Therefore, in the disclosed embodiment, the original historical output data of the photovoltaic power station can be obtained, and the original historical output data can be preprocessed to obtain the historical daily output curve, and then the historical daily output curve is divided into a first category based on the mean clustering algorithm of the first distance index to obtain the corresponding first weather category, and the first weather category includes a high fluctuation category, and then the target historical daily output curve under the high fluctuation category is divided into a second category based on the mean clustering algorithm of the second distance index to obtain the corresponding second weather category, and finally the photovoltaic typical output curves corresponding to the first weather category and the second weather category are extracted respectively. Therefore, the historical daily output curve is subjected to a two-stage clustering process to extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category, so as to accurately extract the photovoltaic typical output curves corresponding to different weather categories.

[0055] Optionally, S110 may specifically include: performing missing value processing, outlier processing and smoothing noise reduction processing on the original historical output data to obtain a historical daily output curve set, wherein the historical daily output curve set includes multiple historical daily output curves, and the historical daily output curve set includes multiple non-intersecting subsets.

[0056] In the disclosed embodiment, the electronic device may perform missing value processing, outlier processing, and smoothing and noise reduction processing on the original historical output data to obtain a historical daily output curve set.

[0057] Optionally, the historical daily effort curve set may include a plurality of the historical daily effort curves, wherein the historical daily effort curve set includes a plurality of mutually non-intersecting subsets.

[0058] Specifically, the electronic device can process missing values ​​of the original historical output data. For example, based on the time resolution of the historical output data and actual needs, an upper limit τ of the number of missing values ​​is set. If the amount of missing data within a day is less than or equal to τ, the missing data is filled by interpolation; if the amount of missing data within a day is greater than τ, all data for that day is removed. The historical output data obtained after the original historical output data Ω0 is processed for missing values ​​is recorded as Ω′0.

[0059] Furthermore, the electronic device performs outlier processing on the historical output data Ω′0. Based on the predetermined abnormal data identification method, the abnormal data of Ω′0 is screened and eliminated, and the data missing after the abnormal data is eliminated is counted. If the missing amount of data within a day is less than or equal to τ, the interpolation method is used to fill the missing data; if the missing amount of data within a day is greater than τ, all data of that day are removed. The historical output data obtained after the outlier processing of Ω′0 is recorded as Ω′0′.

[0060] Furthermore, the electronic device performs smoothing and noise reduction processing on the historical output data Ω′0′. The daily time series output data in Ω′0′ is smoothed and noise-reduced using pre-determined filters such as Savitzky-Golay, and the output data with extremely low output throughout the day are removed to obtain the historical daily output curve set Ω for two-stage clustering analysis. On this basis, Ω can be further divided into non-intersecting subsets according to the season or solar term corresponding to the power generation day, that is, Where Ω m represents the mth subset, and M is the number of subsets.

[0061] Optionally, S120 may specifically include: performing a first category classification process on the historical daily output curves in the target subset of the historical daily output curve set based on a mean clustering algorithm of a first distance index to obtain the corresponding first weather category.

[0062] In an embodiment of the present disclosure, the electronic device may perform a first category classification process on the historical daily output curves in the target subset of the historical daily output curve set based on a mean clustering algorithm of a first distance indicator to obtain the corresponding first weather category.

[0063] Specifically, the electronic device can perform a first category classification process on the target subset of the historical daily effort curve set based on the mean clustering algorithm of the first distance indicator, that is, the target subset Ω of the historical daily effort curve set Ω is classified based on the fuzzy C-means clustering algorithm of the δ distance. m The historical daily output curve is divided into the first category to obtain the corresponding first weather category, which can include non-high fluctuation categories (such as "sunny", "cloudy", "few clouds") and high fluctuation categories.

[0064] Furthermore, the method for extracting typical photovoltaic output curves may also include: iteratively processing the historical daily output curves in the target subset of the historical daily output curve set based on the mean clustering algorithm of the first distance indicator, and determining a cluster center curve set containing multiple first cluster center curves; calculating the intensity score and fluctuation score of each first cluster center curve in the cluster center curve set; and classifying each of the first cluster center curves based on preset classification rules, the intensity score and the fluctuation score to obtain the corresponding first weather category.

[0065] In an embodiment of the present disclosure, the electronic device may iteratively process the historical daily effort curves in a target subset of the historical daily effort curve set based on a mean clustering algorithm of a first distance indicator to determine a cluster center curve set including a plurality of first cluster center curves.

[0066] Specifically, for the target subset Ω of the historical daily force curve set Ωm Each historical daily force curve is called a sample, and N1 is Ω m The total number of samples in .

[0067] S21: Extraction of Ω m The key subset Ω mc , the calculation method is as follows:

[0068]

[0069] Where: vector x i Represents Ω m The i-th sample in x i,t is x i The t-th dimension coordinate of (i.e. the output value at time t), They are Ω m The first quartile, median and third quartile of the t-th dimension coordinate data of all samples, λ is an adjustable parameter.

[0070] S22: Set the number of cluster center curves C1 and the membership termination threshold ε U1 And the loss function termination threshold ε L1 , initialize the iteration count variable s=0.

[0071] S23: Initialize cluster center curve set Calculate the membership matrix U (s) And the loss function value L (s) The steps are as follows:

[0072] S231: Select the first cluster center curve and merge it into V (s) The calculation method is as follows;

[0073]

[0074] Where: vector Represents the first cluster center curve, card(Ω mc ) is Ω mc The cardinality of .

[0075] S232: Calculate Ω mc The delta distance between each sample and the cluster center curve set is calculated as follows:

[0076]

[0077] δ(a,b)=|δ a -δ b |.

[0078] δ a =δ a，1 ·δ a，2 .

[0079]

[0080] F a =w1F a,1 +w2F a,2 +w3F a,3 .

[0081] Where: d1(x i ,V (s) ) is the sample x i and the cluster center curve set V (s) δ(a,b) is the δ distance between curve a and curve b, δ a , δ a,1 , δ a,2 The δ score, intensity score, and fluctuation score of curve a are respectively, a,t is the output value of curve a at time t, T is the number of output values ​​contained in the daily output curve, F a 、F a,1 、F a,2 、F a,3 They are the comprehensive, small, medium and large fluctuation numbers of curve a respectively, and w1, w2 and w3 are the small, medium and large fluctuation coefficients respectively.

[0082] The fluctuation number of curve a is determined as follows:

[0083] (1) Find all extreme values ​​of curve a and record the rth extreme value as q in chronological order a,r (r=1,…,R a ), R a is the total number of extreme values ​​of curve a. Initialize r = 1, F a,1 =F a,2 =F a,3 =0.

[0084] (2) If r = R a +1, stop the calculation; otherwise calculate the rth peak-to-valley difference |q a,r+1 -q a,r |.

[0085] (3) If η 11 P e <|q a,r+1 -q a,r |≤η 12 P e , then F n,1 =F n,1 +1; if η 21 P e <|q a,r+1 -q a,r |≤η 22 Pe , then F n,2 =F n,2 +1; if η 31 P e <|q a,r+1 -q a,r |≤η 32 P e , then F n,3 =F n,3 +1. Among them, P e is the rated output value of the photovoltaic power station, η 12 , η 11 are the upper and lower limit coefficients of small fluctuations, η 22 , η 21 are the upper and lower limit coefficients of the medium fluctuation, η 32 , η 31 They are the upper and lower limit coefficients of large fluctuations respectively.

[0086] (4) Let r = r + 1 and return to step (2).

[0087] S233: In Ω mc Select a sample with the farthest delta distance from the cluster center curve set from the samples as the newly added cluster center curve, and merge it into the cluster center curve set;

[0088] S234: Repeat S232 and S233 until the number of cluster center curves reaches C1.

[0089] S235: Calculate the cluster center curve set V (s) The membership matrix U (s) And the loss function value L (s) , the method is as follows:

[0090]

[0091] Where: The cluster center curve set is V (s) is the membership of the i-th sample to the j-th category; Z is an integer set; h1 is the membership factor; The cluster center curve set is V (s) is the delta distance from the i-th sample to the j-th cluster center curve.

[0092] S24: Based on the membership matrix U (s) Obtain a new cluster center curve set V (s+1) , and calculate the corresponding membership matrix U (s+1) And the loss function value L (s+1) . The new cluster center curve set can be calculated by the following formula:

[0093]

[0094] Where: vector Represents the cluster center curve set V (s+1) The j-th cluster center curve in , its coordinates in each dimension are the output values ​​of the cluster center curve at each moment.

[0095] S25: If or |L (s+1) -L (s) |<ε L1 , then terminate the iteration; otherwise, let s=s+1 and return to S24.

[0096] Furthermore, the electronic device may calculate the intensity score and the fluctuation score of each first cluster center curve in the cluster center curve set, and classify each first cluster center curve based on a preset classification rule, the intensity score and the fluctuation score to obtain the corresponding first weather category.

[0097] Specifically, after the iteration is terminated, the cluster center curve set V is calculated (s) The intensity scores and volatility scores of the C1 cluster center curves included. Based on the characteristics of the pre-given C1 weather categories and conventional logical reasoning, the weather category corresponding to each cluster center curve can be determined. Taking the three weather categories of sunny, cloudy, and high volatility as examples, the cluster center curve with the highest intensity score and the lowest volatility score corresponds to the sunny category; the cluster center curve with an intensity score lower than that of sunny and a non-highest volatility score corresponds to the cloudy category, and the cluster center curve with the highest volatility score corresponds to the high volatility category. According to the membership matrix U (s) For all samples, the category corresponding to the maximum membership degree is taken as the category to which the sample belongs.

[0098] Optionally, the photovoltaic typical output curve extraction method may further include: determining the first cluster center curve under each of the first weather categories as the photovoltaic typical output curve corresponding to each of the first weather categories.

[0099] In the embodiment of the present disclosure, the electronic device may determine that the first cluster center curve under each of the first weather categories is the photovoltaic typical output curve corresponding to each of the first weather categories.

[0100] Specifically, taking the three weather categories of sunny, cloudy, and high volatility as examples, the cluster center curve with the highest intensity score and the lowest volatility score is the typical photovoltaic output curve under the sunny category. The cluster center curve with an intensity score lower than that of sunny day and a non-highest volatility score is the typical photovoltaic output curve under the cloudy category. The cluster center curve with the highest volatility score corresponds to the high volatility category.

[0101] Therefore, based on the δ distance index, the influence of the overall intensity of daily irradiance and time-sharing fluctuation on photovoltaic output is considered at the same time. The clustering method based on the δ distance index can help photovoltaic power stations accurately divide the photovoltaic output curves under weather categories such as sunny days, cloudy days, and high fluctuations, and extract the typical photovoltaic output curves under non-high fluctuation weather categories such as sunny days and cloudy days.

[0102] Optionally, S130 may specifically include: iteratively processing the target historical daily output curve under the high volatility category based on the mean clustering algorithm of the second distance indicator to determine a cluster center curve set including multiple second cluster center curves; calculating the first half-day intensity score, the second half-day intensity score and the skewness score of each second cluster center curve in the cluster center curve set; classifying each of the second cluster center curves based on preset classification rules, the first half-day intensity score, the second half-day intensity score and the skewness score to obtain the corresponding second weather category.

[0103] In an embodiment of the present disclosure, the electronic device may iteratively process the target historical daily effort curve under the high fluctuation category based on the mean clustering algorithm of the second distance indicator to determine a cluster center curve set including multiple second cluster center curves.

[0104] Specifically, each historical daily output curve under the high volatility category is called a sample, the set of these samples is recorded as Y, and N2 is the total number of samples in Y.

[0105] S31: Extract the key subset Y of Y c , the calculation method is as follows:

[0106]

[0107] Where: vector y i represents the i-th sample in the sample set Y, y i,t Yes i The t-th dimension coordinate of (i.e. the output value at time t), are the first quartile, median and third quartile of the t-dimensional coordinate data of all samples of Y, and γ is an adjustable parameter.

[0108] S32: Set the number of cluster center curves C2 and the membership termination threshold ε U2 And the loss function termination threshold ε J2 , initialize the iteration count variable s=0.

[0109] S33: Initialize cluster center curve set Calculate the membership matrix U (s) And the loss function value L (s) The steps are as follows:

[0110] S331: Select the first cluster center curve and merge it into V (s) The calculation method is as follows;

[0111]

[0112] Where: vector Represents the first cluster center curve, card(Y c ) is Y c The cardinality of .

[0113] S332: Calculate Y c The beta distance between each sample and the cluster center curve set is calculated as follows:

[0114]

[0115] β(a,b)=|β a -β b |.

[0116] β a =β a，1 β a，2 β a，3 .

[0117] Where: d2(y i ,V (s) ) is the sample y i and the cluster center curve set V (s) β(a,b) is the β distance between curve a and curve b, β a , β a,1 , β a,2 , β a,3 The beta score, the first half-day strength score, the second half-day strength score and the skewness score of curve a are in turn. f It is the dividing point between the half-day and the previous half-day, and ψ1 and ψ2 are auxiliary calculation quantities.

[0118] S333: In Y c A sample with the farthest β distance from the cluster center curve set is selected from the samples as a newly added cluster center curve, and is incorporated into the cluster center curve set;

[0119] S334: Repeat S332 and S333 until the number of cluster center curves reaches C2.

[0120] S335: Calculate the cluster center curve set V (s) The membership matrix U (s) And the loss function value L (s) , the method is as follows:

[0121]

[0122]

[0123] Where: The cluster center curve set is V (s) The membership degree of the i-th sample to the j-th category when Z is an integer set; h2 is the membership factor; The cluster center curve set is V (s) is the β distance from the i-th sample to the j-th category cluster center curve.

[0124] S34: Based on the membership matrix U (s) Obtain a new cluster center curve set V (s+1) , and calculate the corresponding membership matrix U (s+1) And the loss function value J (s+1) . The new cluster center curve set can be calculated by the following formula:

[0125]

[0126] Where: vector Represents the cluster center curve set V (s+1) The j-th cluster center curve in , its coordinates in each dimension are the output values ​​of the cluster center curve at each moment.

[0127] S35: If or |J (s+1) -J (s) |<ε J2 , the iteration is terminated; otherwise, let s=s+1 and return to S34.

[0128] Furthermore, the electronic device can calculate the first half-day intensity score, the second half-day intensity score and the skewness score of each second cluster center curve in the cluster center curve set, and classify each second cluster center curve based on the preset classification rules, the first half-day intensity score, the second half-day intensity score and the skewness score to obtain the corresponding second weather category.

[0129] Specifically, after the iteration is terminated, the cluster center curve set V is calculated (s)The first half-day intensity score, the second half-day intensity score, and the skewness score of the C2 cluster center curves included can determine the weather category corresponding to each cluster center curve based on the characteristics of the pre-given C2 weather categories and conventional logical reasoning. Taking sunny to cloudy, cloudy, and cloudy to sunny as examples, the cluster center curve with the highest intensity score and the lowest skewness score in the first half of the day corresponds to the sunny to cloudy category, the cluster center curve with the highest intensity score and the highest skewness score in the second half of the day corresponds to the cloudy to sunny category, and the cluster center curve with a skewness score that is neither the highest nor the lowest corresponds to the cloudy category. According to the membership matrix U (s) For all samples, the category corresponding to the maximum membership degree is taken as the category to which the sample belongs.

[0130] Optionally, the photovoltaic typical output curve extraction method may further include: determining the second cluster center curve under each of the second weather categories as the photovoltaic typical output curve corresponding to each of the second weather categories.

[0131] Specifically, taking sunny to cloudy, overcast, and overcast to sunny as examples, the cluster center curve with the highest intensity score and the lowest skewness score in the first half of the day is the typical photovoltaic output curve under the sunny to overcast category, the cluster center curve with the highest intensity score and the highest skewness score in the second half of the day is the typical photovoltaic output curve under the overcast to sunny category, and the cluster center curve with a skewness score that is neither the highest nor the lowest is the typical photovoltaic output curve under the overcast category.

[0132] Therefore, based on the β distance index, the impact of the first half-day irradiance, the second half-day irradiance and the weather changes before and after the first and second half-day on the photovoltaic output is considered simultaneously. The clustering method based on the β distance index can help photovoltaic power stations to further divide the high-volatility weather category samples obtained by clustering in the first stage into subcategories such as sunny to cloudy, cloudy, and cloudy to sunny, and extract the typical photovoltaic output curves under the corresponding categories.

[0133] Figure 2 A schematic structural diagram of a photovoltaic typical output curve extraction device provided in an embodiment of the present disclosure is shown.

[0134] like Figure 2 As shown, the photovoltaic typical output curve extraction device 200 may include a first processing module 210 , a second processing module 220 , a third processing module 230 and a curve extraction module 240 .

[0135] The first processing module 210 can be used to obtain original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain a historical daily output curve.

[0136] The second processing module 220 can be used to perform a first category classification process on the historical daily output curve based on a mean clustering algorithm of a first distance index to obtain a corresponding first weather category, wherein the first weather category includes a high fluctuation category.

[0137] The third processing module 230 can be used to perform a second classification process on the target historical daily output curve under the high fluctuation category based on the mean clustering algorithm of the second distance index to obtain a corresponding second weather category.

[0138] The curve extraction module 240 can be used to extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category respectively.

[0139] Therefore, in the disclosed embodiment, the original historical output data of the photovoltaic power station can be obtained, and the original historical output data can be preprocessed to obtain the historical daily output curve, and then the historical daily output curve is divided into a first category based on the mean clustering algorithm of the first distance index to obtain the corresponding first weather category, and the first weather category includes a high fluctuation category, and then the target historical daily output curve under the high fluctuation category is divided into a second category based on the mean clustering algorithm of the second distance index to obtain the corresponding second weather category, and finally the photovoltaic typical output curves corresponding to the first weather category and the second weather category are extracted respectively. Therefore, the historical daily output curve is subjected to a two-stage clustering process to extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category, so as to accurately extract the photovoltaic typical output curves corresponding to different weather categories.

[0140] In some embodiments of the present disclosure, the first processing module 210 may specifically include a first processing unit.

[0141] The first processing unit can be used to perform missing value processing, outlier processing and smoothing noise reduction processing on the original historical output data to obtain a historical daily output curve set, wherein the historical daily output curve set includes multiple historical daily output curves, and the historical daily output curve set includes multiple non-intersecting subsets.

[0142] In some embodiments of the present disclosure, the second processing module 220 may specifically include a second processing unit.

[0143] The second processing unit can be used to perform first category classification processing on the historical daily output curves in the target subset of the historical daily output curve set based on the mean clustering algorithm of the first distance indicator to obtain the corresponding first weather category.

[0144] In some embodiments of the present disclosure, the second processing module 220 may further specifically include a third processing unit, a first calculating unit and a first dividing unit.

[0145] The third processing unit can be used to iteratively process the historical daily effort curves in the target subset of the historical daily effort curve set based on the mean clustering algorithm of the first distance indicator, and determine a cluster center curve set including multiple first cluster center curves.

[0146] The first calculation unit may be used to calculate the strength score and the fluctuation score of each first cluster center curve in the cluster center curve set.

[0147] The first classification unit may be configured to classify each of the first cluster center curves into categories based on a preset classification rule, the intensity score, and the fluctuation score to obtain the corresponding first weather category.

[0148] In some embodiments of the present disclosure, the photovoltaic typical output curve extraction device 200 may further include a first determination module.

[0149] The first determination module can be used to determine that the first cluster center curve under each of the first weather categories is the photovoltaic typical output curve corresponding to each of the first weather categories.

[0150] In some embodiments of the present disclosure, the third processing module 230 may further specifically include a fourth processing unit, a second calculation unit and a second division unit.

[0151] The fourth processing unit can be used to iteratively process the target historical daily effort curve under the high fluctuation category based on the mean clustering algorithm of the second distance indicator, and determine a cluster center curve set including multiple second cluster center curves.

[0152] The second calculation unit can be used to calculate the first half-day intensity score, the second half-day intensity score and the skewness score of each second cluster center curve in the cluster center curve set.

[0153] The second classification unit can be used to classify each of the second cluster center curves based on a preset classification rule, a first half-day intensity score, a second half-day intensity score and a skewness score to obtain the corresponding second weather category.

[0154] In some embodiments of the present disclosure, the photovoltaic typical output curve extraction device 200 may further include a second determination module.

[0155] The second determination module may be used to determine that the second cluster center curve under each of the second weather categories is the photovoltaic typical output curve corresponding to each of the second weather categories.

[0156] It should be noted that Figure 2 The photovoltaic typical output curve extraction device 200 shown can be executed Figure 1 The various steps in the method embodiment shown in the figure are implemented Figure 1 The various processes and effects in the method embodiment shown are not described in detail here.

[0157] Figure 3 A schematic diagram of the structure of a photovoltaic typical output curve extraction device provided in an embodiment of the present disclosure is shown.

[0158] In some embodiments of the present disclosure, Figure 3 The photovoltaic typical output curve extraction device shown may be an electronic device. Specifically, the electronic device may include but is not limited to devices such as computer devices, cloud servers or cloud server clusters.

[0159] like Figure 3 As shown, the photovoltaic typical output curve extraction device may include a processor 301 and a memory 302 storing computer program instructions.

[0160] Specifically, the processor 301 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0161] The memory 302 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 302 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 302 may be inside or outside the integrated gateway device. In a particular embodiment, the memory 302 is a non-volatile solid-state memory. In a particular embodiment, the memory 302 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0162] The processor 301 reads and executes the computer program instructions stored in the memory 302 to perform the steps of the photovoltaic typical output curve extraction method provided in the embodiment of the present disclosure.

[0163] In one example, the photovoltaic typical output curve extraction device may further include a transceiver 303 and a bus 304. Figure 3 As shown, the processor 301, the memory 302 and the transceiver 303 are connected via a bus 304 and communicate with each other.

[0164] The bus 304 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a Memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 304 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0165] The embodiments of the present disclosure also provide a computer-readable storage medium, which may store a computer program. When the computer program is executed by a processor, the processor implements the photovoltaic typical output curve extraction method provided by the embodiments of the present disclosure.

[0166] The above-mentioned storage medium may, for example, include a memory 302 of computer program instructions, and the above-mentioned instructions may be executed by the processor 301 of the photovoltaic typical output curve extraction device to complete the photovoltaic typical output curve extraction method provided by the embodiment of the present disclosure. Optionally, the storage medium may be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium may be a ROM, a random access memory (Random Access Memory, RAM), a compact disc read-only memory (Compact Disc ROM, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0167] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0168] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for extracting a typical photovoltaic output curve, characterized in that: include: Acquire original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain a historical daily output curve; Performing a first category classification process on the historical daily output curve based on a mean value clustering algorithm of a first distance index to obtain a corresponding first weather category, wherein the first weather category includes a high fluctuation category; Based on the mean value clustering algorithm of the second distance index, the target historical daily output curve under the high fluctuation category is divided into the second category to obtain the corresponding second weather category; The photovoltaic typical output curves corresponding to the first weather category and the second weather category are extracted respectively.

2. The method according to claim 1, characterized in that The data preprocessing of the original historical output data to obtain the historical daily output curve includes: The original historical output data is processed for missing values, outliers and smoothing noise reduction to obtain a historical daily output curve set, wherein the historical daily output curve set includes a plurality of the historical daily output curves, and the historical daily output curve set includes a plurality of mutually non-intersecting subsets.

3. The method according to claim 2, characterized in that The mean value clustering algorithm based on the first distance index performs a first category classification process on the historical daily output curve to obtain a corresponding first weather category, including: A mean clustering algorithm based on a first distance index is used to perform a first category classification process on the historical daily output curves in a target subset of the historical daily output curve set to obtain the corresponding first weather category.

4. The method according to claim 3, characterized in that The mean value clustering algorithm based on the first distance index performs a first category classification process on the historical daily output curves in the target subset of the historical daily output curve set to obtain the corresponding first weather category, including: Iteratively processing the historical daily effort curves in the target subset of the historical daily effort curve set using a mean value clustering algorithm based on a first distance indicator to determine a cluster center curve set including a plurality of first cluster center curves; Calculating the intensity score and the fluctuation score of each first cluster center curve in the cluster center curve set; Each of the first cluster center curves is classified based on a preset classification rule, the intensity score, and the fluctuation score to obtain the corresponding first weather category.

5. The method according to claim 4, characterized in that The method further comprises: The first cluster center curve under each of the first weather categories is determined as the photovoltaic typical output curve corresponding to each of the first weather categories.

6. The method according to claim 1, characterized in that The mean value clustering algorithm based on the second distance index performs a second classification process on the target historical daily output curve under the high fluctuation category to obtain a corresponding second weather category, including: Iteratively process the target historical daily effort curve under the high fluctuation category using a mean value clustering algorithm based on a second distance indicator to determine a cluster center curve set including a plurality of second cluster center curves; Calculating the first half-day intensity score, the second half-day intensity score and the skewness score of each second cluster center curve in the cluster center curve set; Based on the preset classification rules, the first half-day intensity score, the second half-day intensity score and the skewness score, each of the second cluster center curves is classified to obtain the corresponding second weather category.

7. The method according to claim 6, characterized in that The method further comprises: The second cluster center curve under each of the second weather categories is determined as the photovoltaic typical output curve corresponding to each of the second weather categories.

8. A photovoltaic typical output curve extraction device, characterized in that: include: The first processing module is used to obtain the original historical output data of the photovoltaic power station, and perform data preprocessing on the original historical output data to obtain a historical daily output curve; A second processing module is used to perform a first category classification process on the historical daily output curve based on a mean value clustering algorithm of a first distance indicator to obtain a corresponding first weather category, wherein the first weather category includes a high fluctuation category; A third processing module is used to perform a second classification process on the target historical daily output curve under the high fluctuation category based on a mean value clustering algorithm of a second distance indicator to obtain a corresponding second weather category; The curve extraction module is used to extract the photovoltaic typical output curves corresponding to the first weather category and the second weather category respectively.

9. A photovoltaic typical output curve extraction device, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the photovoltaic typical output curve extraction method described in any one of claims 1 to 7.

10. A non-volatile computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the processor implements the photovoltaic typical output curve extraction method described in any one of claims 1 to 7.

Images