Data processing method and device of photovoltaic power station, electronic equipment and storage medium

Abstract

The invention relates to the technical field of data processing, in particular to a data processing method and device for a photovoltaic power station, electronic equipment and a storage medium, and the method comprises the steps: obtaining time series data of the photovoltaic power station; and performing multi-dimensional data clustering on the time sequence data to obtain multi-dimensional similarity measurement of the photovoltaic power station, and generating a target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station. Therefore, multiple dimensions including the power dimension, the energy dimension and the volatility of the photovoltaic power station are considered, the representative photovoltaic scene is effectively extracted, the data volume is reduced, the data analysis and processing efficiency is improved, and the method is suitable for being popularized and applied. And evaluation is carried out by using indexes such as random power flow, annual energy output, photovoltaic daily energy output curve shape similarity and photovoltaic power generation peak period frequency value of photovoltaic grid connection, the accuracy of photovoltaic scene extraction is verified from two aspects of a clustering algorithm and power, and the reliability and practicability are higher.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data processing method, device, electronic equipment and storage medium for a photovoltaic power station. Background Art

[0002] With the development of solar power generation technology and the rapid growth of photovoltaic scale, the effective use of photovoltaic historical data is crucial to improving the efficiency of photovoltaic grid-connected, formulating reasonable inverter control strategies, implementing uncertainty scheduling and planning, and accurately simulating and predicting photovoltaic power generation.

[0003] However, faced with massive amounts of historical photovoltaic data, its processing faces two major challenges: first, the high repetitiveness and regularity of the data leads to a huge waste of storage and computing resources; second, the existing technology for extracting effective data from scenarios cannot truly reflect the actual photovoltaic power generation situation, which in turn affects the accuracy and reliability of power system planning. Summary of the invention

[0004] The present application provides a data processing method, device, electronic device and storage medium for a photovoltaic power station to solve the problems that the relevant technology cannot truly reflect the actual photovoltaic power generation situation, affects the accuracy and reliability of photovoltaic power station planning, and has a large amount of data processing.

[0005] A first aspect of the present application provides a data processing method for a photovoltaic power station, comprising the following steps: acquiring time series data of the photovoltaic power station; performing multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station, wherein the multi-dimensionality is a plurality of power dimensions, energy dimensions, and volatility dimensions of the photovoltaic power station; and generating a target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station.

[0006] Optionally, multi-dimensional data clustering is performed on the time series data to obtain a multi-dimensional similarity measure of the photovoltaic power station, including: pre-clustering the time series data to obtain a target number of clusters; inputting the data of the target number of clusters into a clustering model, and the clustering model outputs a multi-dimensional similarity measure of the photovoltaic power station, wherein the clustering model calculates the similarity measure, power size similarity, mutual correlation sequence length and fluctuation position similarity of any two power time series in the data of the target number of clusters, and generates a multi-dimensional similarity measure based on the similarity measure, power size similarity, mutual correlation sequence length and fluctuation position similarity.

[0007] Optionally, the clustering model is:

[0008] LESD(x i ,x j ,W)=w1L(x i ,x j)+w2E(x i ,x j )+w3S(x i ,x j )+w4δ(x i ,x j )

[0009] Among them, w1, w2, w3 and w4 represent weight coefficients, forming the vector W∈R 4×1 , x i and x j Represents a power time series of length L, x it , x jt Respectively represent the power time series x at time t i and x j Features;

[0010] E(x i ,x j ) represents x i and x j The similarity of the electric quantity;

[0011] CC ω (x i ,x j )=R ω-L (x i ,x j ),ω∈{1,2,K,2L-1},

[0012] CC w (x i ,x j ) represents x i and x j The cross-correlation sequence of s represents the power time series x j The moving step length, s∈[-L,L]; w is the cross-correlation sequence CC w (x i ,x j ) position subscript;

[0013] S(x i ,x j )=1-Max(NCC c (x i ,x j )),S(x i ,x j ) represents x i With x j The morphological trend similarity distance has a range of [0,2];

[0014] Represents the sequence x j Slide to and x i The displacement when the morphological trend is most similar, δ(x i ,x j ) indicates the similarity of the fluctuations of the two power time series.

[0015] Optionally, before inputting the data of the target number of clusters into the clustering model, the method further includes: obtaining neighborhood probabilities of the power time series and other sequences; and determining the neighborhood sequence of the power time series according to the neighborhood probabilities.

[0016] Optionally, generating a target photovoltaic scene of the photovoltaic power station according to the multidimensional similarity measurement of the photovoltaic power station includes: extracting a morphologically representative scene from the multidimensional similarity measurement of the photovoltaic power station; and amplifying the morphologically representative scene to obtain the target photovoltaic scene of the photovoltaic power station.

[0017] The target photovoltaic scenarios are:

[0018]

[0019] Among them, N g is the number of power time series of the gth category; x g,i,t is the characteristic value of the i-th power time series in the g-th category at time t; is the representative scene of the g-th type of morphology, μ g,t * is the characteristic value of the representative scene of the g-th type of morphology at time t, P g The original sequence set representing the cluster where the representative scene to be sought belongs.

[0020] Optionally, before performing multi-dimensional data clustering on the time series data to obtain the multi-dimensional similarity measurement of the photovoltaic power station, the method further includes: performing standardization processing on the time series data.

[0021] Optionally, after generating the target photovoltaic scenario of the photovoltaic power station according to the multidimensional similarity measurement of the photovoltaic power station, it also includes: obtaining multiple indicator data of random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period; based on the multiple indicator data of random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period, evaluating the accuracy level of the target photovoltaic scenario.

[0022] A second aspect of the present application provides a data processing device for a photovoltaic power station, including: an acquisition module for acquiring time series data of the photovoltaic power station; a clustering module for performing multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station, wherein the multi-dimensionality is a plurality of power dimensions, energy dimensions and volatility dimensions of the photovoltaic power station; and a generation module for generating a target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station.

[0023] Optionally, the clustering module is further used to pre-cluster the time series data to obtain a target number of clusters; the data of the target number of clusters is input into the clustering model, and the clustering model outputs a multidimensional similarity measure of the photovoltaic power station, wherein the clustering model calculates the similarity measure, the similarity of the amount of electricity, the length of the mutual correlation sequence and the similarity of the fluctuation position of any two power time series in the data of the target number of clusters, and generates a multidimensional similarity measure based on the similarity measure, the similarity of the amount of electricity, the length of the mutual correlation sequence and the similarity of the fluctuation position.

[0024] Optionally, the clustering model is:

[0025] LESD(x i ,x j ,W)=w1L(x i ,x j )+w2E(x i ,x j )+w3S(x i ,x j )+w4δ(x i ,x j )

[0026] Among them, w1, w2, w3 and w4 represent weight coefficients, forming the vector W∈R 4×1 , x i and x j Represents a power time series of length L, x it , x jt Respectively represent the power time series x at time t i and x j Features;

[0027] E(x i ,x j ) represents x i and x j The similarity of the electric quantity;

[0028] CC ω (x i ,x j )=R ω-L(x i ,x j ),ω∈{1,2,K,2L-1},

[0029] CC w (x i ,x j ) represents x i and x j The cross-correlation sequence of s represents the power time series x j The moving step length, s∈[-L,L]; w is the cross-correlation sequence CC w (x i ,x j ) position subscript;

[0030] S(x i ,x j )=1-Max(NCC c (x i ,x j )),S(x i ,x j ) represents x i With x j The morphological trend similarity distance has a range of [0,2];

[0031] Represents the sequence x j Slide to and x i The displacement when the morphological trend is most similar, δ(x i ,x j ) indicates the similarity of the fluctuations of the two power time series.

[0032] Optionally, the data processing device of the photovoltaic power station also includes: a determination module, which is used to obtain the neighborhood probability of the power time series and other sequences before inputting the data of the target number of clusters into the clustering model; and determine the neighborhood sequence of the power time series according to the neighborhood probability.

[0033] Optionally, the generation module is further used to extract a morphologically representative scene from the multidimensional similarity measurement of the photovoltaic power station; the morphologically representative scene is amplified to obtain a target photovoltaic scene of the photovoltaic power station, wherein the target photovoltaic scene is:

[0034]

[0035] Among them, N g is the number of power time series of the gth category; x g,i,t is the characteristic value of the i-th power time series in the g-th category at time t; is the representative scene of the g-th type of morphology, μ g,t *is the characteristic value of the representative scene of the g-th type of morphology at time t, P g The original sequence set representing the cluster where the representative scene to be sought belongs.

[0036] Optionally, the data processing device of the photovoltaic power station further includes: a processing module, which is used to perform standardization processing on the time series data before performing multi-dimensional data clustering on the time series data to obtain the multi-dimensional similarity measurement of the photovoltaic power station.

[0037] Optionally, the data processing device of the photovoltaic power station also includes: an evaluation module, which is used to obtain multiple indicator data of random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape, and frequency value of photovoltaic power generation peak period after generating the target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station; and evaluate the accuracy level of the target photovoltaic scene based on the multiple indicator data of random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape, and frequency value of photovoltaic power generation peak period.

[0038] The third aspect of the present application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a data processing method for a photovoltaic power station as described in the above embodiment.

[0039] A fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the data processing method for a photovoltaic power station as described in the above embodiment.

[0040] Therefore, this application includes the following beneficial effects:

[0041] The embodiment of the present application obtains a multidimensional similarity measurement of a photovoltaic power station by performing multidimensional data clustering on time series data, which can more comprehensively reflect the actual characteristics of photovoltaic power generation. Based on the multidimensional similarity measurement of photovoltaic power stations, representative photovoltaic scenes are effectively extracted, the amount of data is reduced, the efficiency of data analysis and processing is improved, and by evaluating the target photovoltaic scene, a reliable basis is provided for the scheduling, control strategy formulation and future scenario simulation of the power system. Thus, the problems that the relevant technology cannot truly reflect the actual photovoltaic power generation situation, affect the accuracy and reliability of photovoltaic power station planning, and have a large amount of data processing are solved.

[0042] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0044] Figure 1 A flowchart of a data processing method for a photovoltaic power station provided according to an embodiment of the present application;

[0045] Figure 2 A schematic diagram of a process of obtaining a representative photovoltaic scene through an optimization algorithm according to an embodiment of the present application;

[0046] Figure 3 A schematic diagram of a photovoltaic representative scenario result of an embodiment of the present application;

[0047] Figure 4 A schematic diagram showing comparison of probability power flow results of an IEEE 69 bus power distribution system according to an embodiment of the present application;

[0048] Figure 5 A schematic diagram showing a comparison between the total power of a photovoltaic power station according to an embodiment of the present application and the power area after clustering using representative scenes to replace corresponding original data points;

[0049] Figure 6 A schematic diagram showing the distribution of the relative frequency of the peak moments of a representative scene and the relative frequency of the peak moments of an original scene according to an embodiment of the present application;

[0050] Figure 7 is a schematic structural diagram of a data processing device for a photovoltaic power station according to an embodiment of the present application;

[0051] Figure 8 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0053] The data processing method, device, electronic device and storage medium of the photovoltaic power station of the embodiment of the present application are described below with reference to the accompanying drawings. In response to the problems mentioned in the above background technology, the present application provides a data processing method for a photovoltaic power station, which more fully considers the characteristics of electricity and power time series, so that the results are more accurate and have practical significance. At the same time, a method for comprehensively evaluating the accuracy of photovoltaic scene extraction is proposed, which combines probabilistic flow analysis, electricity characteristics, morphological similarity verification and relative frequency values ​​at peak moments, and is evaluated through dual verification indicators, namely mathematical theory verification and actual business needs of the power system, to ensure the accuracy and practicality of the results. In this method, a multi-dimensional similarity measurement of the photovoltaic power station is obtained by multi-dimensional data clustering of time series data, which can more comprehensively reflect the actual characteristics of photovoltaic power generation, effectively extract representative photovoltaic scenes, reduce the amount of data, and improve the efficiency of data analysis and processing.

[0054] Specifically, Figure 1 A schematic flow chart of a data processing method for a photovoltaic power station provided in an embodiment of the present application.

[0055] like Figure 1 As shown, the data processing method of the photovoltaic power station includes the following steps:

[0056] In step S101 , time series data of a photovoltaic power station is obtained.

[0057] It can be understood that the embodiment of the present application can obtain the photovoltaic power generation time series data to be clustered through data collection. For example, the photovoltaic power generation time series data collection range can be set to 1 year, and the collection frequency can be set to 1h / time. In the actual execution process, the collected series of photovoltaic power generation data are arranged in chronological order to form a photovoltaic power generation time series data to be clustered X=(x1, x2,…, x d ).

[0058] In step S102, multi-dimensional data clustering is performed on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station.

[0059] Among them, the multiple dimensions are multiple dimensions of power, energy and volatility of photovoltaic power stations.

[0060] In an embodiment of the present application, before performing multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station, the embodiment of the present application can perform Max-Min normalization processing on the time series data, which can eliminate the dimensional differences between different dimensions (such as different photovoltaic power stations or different power generation levels) and obtain time series data after normalization processing of different dimensions.

[0061] In one embodiment of the present application, multi-dimensional data clustering is performed on time series data to obtain a multi-dimensional similarity measure of a photovoltaic power station, including: pre-clustering the time series data to obtain a target number of clusters; inputting the data of the target number of clusters into a clustering model, and the clustering model outputs a multi-dimensional similarity measure of the photovoltaic power station, wherein the clustering model calculates the similarity measure, the similarity of the amount of electricity, the length of the mutual correlation sequence, and the similarity of the fluctuation position of any two power time series in the data of the target number of clusters, and generates a multi-dimensional similarity measure based on the similarity measure, the similarity of the amount of electricity, the length of the mutual correlation sequence, and the similarity of the fluctuation position.

[0062] It is understandable that the embodiment of the present application can perform pre-clustering (such as using K-means clustering) on ​​the standardized time series data obtained in the above steps, and the number of clusters can be 2 to 100. At the same time, in order to avoid the randomness of initialization, the embodiment of the present application can be run 100 times, and the best result is selected as the final clustering result to ensure that the final clustering result is relatively stable and representative. In the actual execution process, the embodiment of the present application can select the number of clusters corresponding to the fastest growth rate, and can select a smaller number of clusters to represent most of the overall information of the data, while saving the computing cost to a certain extent.

[0063] Furthermore, the embodiments of the present application can establish a clustering model that takes into account a multidimensional similarity measurement of power, energy, and volatility, input the data of the target number of clusters into the clustering model, and the clustering model outputs a multidimensional similarity measurement of photovoltaic power stations. The clustering model can measure the similarity of photovoltaic power generation data in four dimensions: distance size, power size, morphological trend, and fluctuation position, thereby improving the accuracy of photovoltaic scene extraction.

[0064] Assume that two power time series x are both of length L i =(x i1 ,x i2 ,…,x iL ) T and x j =(x j1 ,x j2 ,…,x jL ) T .

[0065] The distance similarity metric of the present application embodiment can be measured by the Euclidean distance between two power time series x. i and x j The smaller the Euclidean distance, the more similar the two power time series are. The similarity metric L(x i ,x j ) is defined as follows:

[0066]

[0067] Among them, x it , x jt Represent the power time series x at time t respectively i and x j The characteristics of L(x i ,x j ) selects the square form of Euclidean distance, which is easy to derive and optimize. L(x i ,x j )The smaller the value, the more similar the changes in the characteristic values ​​of the power time series are; the larger the value, the greater the changes in the characteristic values ​​of the power time series are.

[0068] In the embodiment of the present application, the power difference represents the shaded area of ​​the two power time series and the coordinate axis. If the two power time series curves are similar, the power size of the two must be similar. i and x j The similarity of the electric quantity E(x i ,x j ) is defined as follows:

[0069]

[0070] Among them, x i and x j Represent the power time series x at time t respectively i and x j Features, and Represents the sum of the eigenvalues ​​of each power time series, and the standardized metric value is always greater than or equal to 1. i ,x j )The smaller the value, the more similar the two charges are; the larger the value, the greater the difference in charge between the two charges.

[0071] The morphological trend of the embodiment of the present application represents the characteristic conditions in different time periods, so the morphological trends of two power time series under similar weather conditions should be similar. The method for calculating the cross-correlation of the series is as follows: displacement invariance ensures that the translation of the power time series on the time axis will not affect the results of its correlation analysis: when calculating the cross-correlation of two power time series, one sequence is fixed and the other sequence is slid in the specified direction and step size. i and x j The cross-correlation sequence is CC w (x i ,x j )=(c1,…,c w ), the length of this sequence is (2L-1), and is defined as follows:

[0072] CC ω(x i ,x j )=R ω-L (x i ,x j ),ω∈{1,2,K,2L-1},

[0073]

[0074] Where s represents the power time series x j The moving step length, s∈[-L,L]; w is the sequence CC w (x i ,x j ) position subscript. Through the sequence CC w (x i ,x j ) can get the position w corresponding to the maximum cross-correlation value max According to this position, we can get the power time series x when the cross-correlation value is maximum j Relative to the power time series x i The alignment sequence x j '.

[0075] It can be understood that in order to improve the comparability of the cross-correlation sequence data, the coefficient normalization method is used to process the cross-correlation sequence so that the zero-lag autocorrelation coefficient is equal to 1. The coefficient normalization formula is as follows:

[0076]

[0077] Among them, NCC c (x i ,x j ) is the cross-correlation sequence after coefficient normalization, and its value range is [-1,1]; R0(x i ,x i ) and R0(x j ,x j ) are the power time series x i and x j When NCC c (x i ,x j ) is 1 or -1, the two sequences are positively or negatively correlated, so NCC c (x i ,x j ) value, the more similar the two sequences are. For the convenience of calculation, define the sequence x i With x j The morphological trend similarity distance is:

[0078] S(x i ,x j)=1-Max(NCC c (x i ,x j )),

[0079] Among them, S(x i ,x j ) has a value range of [0,2], where 0 represents the sequence x i With x j The shape trends of are exactly the same, and 2 represents the exact opposite.

[0080] The similarity of the fluctuation position in the embodiment of the present application is an important factor in the similarity of the power time series curves. Since the cross-correlation is to calculate the morphological trend distance of the two power time series under the condition of considering the displacement invariance, that is, the sequence x i and the sequence x j The morphological trend distance is given by the sequence x i and the sequence x j 'Calculated, thus ignoring the deviation of the fluctuation positions of the two. Therefore, the fluctuation position similarity δ of the two sequences is defined to reflect the difference in the occurrence time of the fluctuations of the two sequences. The formula is as follows:

[0081]

[0082] in, Represents the sequence x j Slide to and x i The displacement when the morphological trend is most similar, δ(x i ,x j )The smaller the value, the more similar the fluctuation positions of the two power time series are; the larger the value, the greater the difference in the fluctuation positions of the two power time series.

[0083] Finally, in order to reflect the multidimensional similarity between power time series, the multidimensional similarity measure LESD (i.e., clustering model) is obtained based on the above four different unit type distances:

[0084] LESD(x i ,x j ,W)=w1L(x i ,x j )+w2E(x i ,x j )+w3S(x i ,x j )+w4δ(x i ,x j )

[0085] Among them, w1, w2, w3 and w4 represent weight coefficients, forming the vector W∈R 4×1 , define constraint 0 <w1,w2,w3,w4<1,W T1=1.

[0086] In the embodiment of the present application, before inputting the data of the target number of clusters into the clustering model, it also includes: obtaining the neighborhood probabilities of the power time series and other sequences; and determining the neighborhood sequence of the power time series according to the neighborhood probabilities.

[0087] In order to fully explore the local connectivity between power time series, the embodiment of the present application can introduce a probability neighborhood for each power time series. i , in all power time series x i With any other data x j The neighborhood probabilities between them are all obtained through a ij The smaller the multidimensional similarity measure, the greater the similarity a ij Therefore, the similarity a ij It can be solved as follows: for each power time series x i :

[0088]

[0089] Among them, a ij is the power time series x i and x j The connection probability between them, 0≤a ij ≤1 and vector a i ∈R d×1 The jth element of is a ij , a i T 1=1, 1 is a column vector whose elements are all 1, a ij Constitute A∈R d×d , d represents the number of days, and L represents the power values ​​of the five photovoltaic stations at 24 moments. Multidimensional similarity metric LESD(x i ,x j ,W) is smaller, the adjacent probability a ij The bigger.

[0090] The present application embodiment introduces Get a ij , x i The neighborhood of can be any power time series other than itself.

[0091] In step S103, a target photovoltaic scene of the photovoltaic power station is generated according to the multi-dimensional similarity measurement of the photovoltaic power station.

[0092] In one embodiment of the present application, a target photovoltaic scene of a photovoltaic power station is generated according to a multidimensional similarity measurement of the photovoltaic power station, including: extracting a morphologically representative scene from the multidimensional similarity measurement of the photovoltaic power station; and amplifying the morphologically representative scene to obtain the target photovoltaic scene of the photovoltaic power station.

[0093] The embodiment of the present application can introduce a regularization parameter γ. For any two power time series, a ij This can be derived from the following questions:

[0094]

[0095] The second term in the above formula is regularization, and γ is the regularization parameter. There is only one connected component in the solution of the above formula, and the ideal neighborhood allocation of the algorithm is to have c connected branches, so the following properties in graph theory are used:

[0096] Theorem 1 (K-connected branches): The Laplace matrix L A The multiplicity c of the eigenvalue 0 is equal to the number of connected components in the graph with the similarity matrix A.

[0097]

[0098] Among them, L A =D A -(A T +A) / 2 degree matrix D A ∈R d×d The i-th diagonal element of is Σ j (a ij +a ji ) / 2, {α1,α2,…,α d} is L A The characteristic value of .

[0099] It can be understood that according to Theorem 1, the Laplace matrix rank constraint of the similarity matrix A is imposed on the above formula to ensure that the neighborhood allocation of A is ideal, so that the photovoltaic output data set can be divided into c clusters. The representative scene extraction model after adding this constraint can be expressed as follows:

[0100]

[0101] Theorem 2 (Properties of the Laplace Matrix): The Laplace matrix L A It has semi-positive definiteness, that is, L A All eigenvalues ​​of are non-negative.

[0102] It can be understood that, in the embodiment of the present application, assuming that the Laplace matrix L A The gth smallest eigenvalue of g (LA )≥0, from Theorem 2 we know that σ g (L A )≥0, on this basis, we introduce a sufficiently large parameter λ, combined with Theorem 1, to satisfy the Laplace matrix rank constraint rank(L A )=dc, so the above formula is converted into the following form:

[0103]

[0104] Theorem 3 (Ky Fan Theorem): For a matrix F∈R d×c , there is the following minimization problem:

[0105]

[0106] Among them, F∈R d×c The i-th row of is the function value assigned to each data point Tr(·) is the trace of the matrix.

[0107] It can be understood that according to Theorem 3, for any Laplace matrix L A and L A The sum of the first c smallest eigenvalues ​​of can be obtained as follows:

[0108]

[0109] Finally, the following representative scene extraction model can be obtained:

[0110]

[0111] Furthermore, the embodiments of the present application can obtain representative photovoltaic scenes through an alternating optimization algorithm and a two-stage optimization extraction algorithm.

[0112] In the actual implementation process, the representative scenes are divided into two stages. In the first stage, the representative scenes μ of the cluster morphology are extracted by using the optimization idea. g * In the second stage, the same magnification method is used to enlarge it to the actual representative scene μ g In the first stage, the representative scene of the cluster is the power time series curve with the closest multidimensional similarity distance to other curves in the cluster. The representative scene μ g * It is expressed as follows:

[0113]

[0114] Among them, P g The original sequence set representing the cluster where the representative scene to be sought belongs.

[0115] In the second stage, the same ratio magnification method is used to enlarge it to the actual representative scene μ g , set to the original cluster power time series mean and morphological representative scene μ g * The ratio of the mean, the actual representative scenario of each photovoltaic power station μ g It is expressed as follows:

[0116]

[0117] Among them, N g is the number of power time series of the gth category; x g,i,t is the characteristic value of the i-th power time series in the g-th category at time t; μ g,t * is the characteristic value of the representative scene of the g-th type of morphology at time t, P g The original sequence set representing the cluster where the representative scene to be sought belongs.

[0118] In summary, the embodiment of the present application measures the adjacency probability of data points based on multi-dimensional similarity distance, and obtains the actual representative scene of each photovoltaic power station through a two-stage optimization extraction method, thereby completing the division and extraction of representative photovoltaic output scenes.

[0119] The following embodiment of the present application can be used as a specific embodiment to describe the content of the data processing method of the photovoltaic power station in detail, such as Figure 2 As shown, the following steps are included:

[0120] Step S201: Obtain the photovoltaic power generation time series data to be clustered X = (x1, x2, ..., x d ), the length of the time series is L.

[0121] Specifically, the photovoltaic power generation time series data X to be clustered is obtained, with the entire year of year B as the collection range, the collection frequency is 1 hour / time, there are 24 sampling data points every day, and a total of 365 data.

[0122] Step S202: Perform Max-Min normalization processing on X.

[0123] S203: Perform K-means pre-clustering on the standardized X, with the number of clusters ranging from 2 to 100. In order to avoid the randomness of initialization, the process is repeated 100 times to obtain a stable clustering result, and the number of clusters corresponding to the fastest growth rate is selected.

[0124] S204: Establish a clustering model and perform optimization and solution to obtain a representative photovoltaic scenario.

[0125] Specifically, the proposed clustering model is optimized and solved by an alternating optimization algorithm including linear optimization solution to obtain W, Lagrange multiplier method to obtain A and eigenvalue decomposition F until the model converges; and

[0126] Through a two-stage optimization extraction algorithm, that is, firstly, the morphological centroid is obtained by minimizing the distance between the data points in the class and the class center, and then the actual centroid, that is, the representative photovoltaic scene, is obtained by the same-multiple magnification method. In order to more intuitively and clearly demonstrate the proposed method, a visualization method of representative photovoltaic scenes is provided, such as Figure 3 As shown, a schematic diagram of the results of representative photovoltaic scenarios of an embodiment of the present application is provided, and the proportion results of various scenarios are given as shown in Table 1.

[0127] Table 1

[0128]

[0129] Furthermore, after generating the target photovoltaic scenario of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station, it also includes: obtaining multiple indicator data of random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period; based on the multiple indicator data of random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period, evaluating the accuracy level of the target photovoltaic scenario.

[0130] Specifically, if Figure 2 In step S205 shown, the embodiment of the present application uses power flow analysis, annual electricity, peak time statistics, and morphological similarity analysis as evaluation indicators to compare the effectiveness of the embodiment of the present application with other related technologies. By using indicators such as random power flow of photovoltaic grid-connected, annual power generation, similarity of photovoltaic daily power generation curve shape, and frequency value of photovoltaic power generation peak period to evaluate, the accuracy of representative photovoltaic scene extraction is verified from the two perspectives of clustering algorithm and power, and the reliability and practicality are stronger.

[0131] Specifically, the peak hour relative frequency is calculated as follows:

[0132] Calculate the peak moment frequency of each category after standardization: For each category in the photovoltaic power generation data, calculate the peak moment frequency per hour, that is, count the time at which the maximum power of this category is reached every day:

[0133]

[0134] Among them, H t,g is the number of peaks of category g in the tth hour, N g is the number of data points in category g.

[0135] For example, in order to verify the correctness of representative photovoltaic power generation scenarios in the power system, the MATPOWER software package was used. The four comparison algorithms include Sparse Subspace Clustering SSC (Sparse Subspace Clustering), Constrained Laplacian Rank Based Upon the L2-norm CLR-L2 (Constrained Laplacian Rank Based Upon the L2-norm), Constrained Laplacian Rank Based Upon the L1-norm CLR-L1 (Constrained Laplacian Rank Based Upon the L1-norm) and Shared Weight Graph Clustering SWGC (Shared Weight Graph Clustering). To avoid the randomness of SSC, the results are the average values ​​of 100 runs. Specifically, the photovoltaic power stations are integrated into nodes 2, 22, 33, 18 and 25 of the case69 test case in turn, as shown in Figure 2. Figure 4 As shown, a schematic diagram of the probabilistic power flow results of an IEEE 69 bus power distribution system according to an embodiment of the present application is shown.

[0136] Depend on Figure 4 It can be seen that in terms of the mean, compared with the other five related technologies, the model proposed in the embodiment of the present application is closest to the true value; in terms of standard deviation, it is also very close to the original real scene. The comparison results of the PPF results show that the proposed model has good performance. Therefore, compared with the related technologies, the algorithm of the embodiment of the present application is better than the comparison algorithm.

[0137] For example, in order to verify that the representative photovoltaic power generation scenario maintains the power characteristics of the original scenario, the original power generation data points are replaced with representative scenarios according to the representative scenarios extracted by the model in this paper to calculate the total power after clustering. The total power of the photovoltaic power station is visually compared with the power area after clustering using representative scenarios to replace the corresponding original data points. The results are as follows: Figure 5 As shown in the figure, the x-axis represents the output time of 365 days, the y-axis represents the power, the curve represents the daily power generation of the original data, and the bar chart represents the daily power generation of the typical scene.

[0138] As can be seen from the figure, the daily power generation of the representative scenes extracted by the representative scene extraction model of the embodiment of the present application is mostly consistent with the daily power generation of the original data. From the observation of the figure, it can be seen that the area of ​​the curve is equal to the total area of ​​the bar chart. The cumulative electricity E obtained by calculating the representative scene extraction model proposed in this article is equal to the original total electricity. This shows that the photovoltaic representative scenes extracted by the embodiment of the present application better reflect the total electricity of the original scene set.

[0139] For example, in order to verify that the representative photovoltaic power generation scenario maintains the volatility characteristics of the original scenario, the power generation of the five photovoltaic power stations at each time is summed up, and the time when the maximum power is reached every day is the peak time of the day. The relative frequency of the peak time of the representative scenario is compared with the relative frequency of the peak time of the original scenario. The results of this technology are as follows: Figure 6 As shown in the figure, the x-axis represents the 13 clusters, and the y-axis represents the peak time of each cluster. The larger the blue circle, the greater the probability that the original scene of this type is the peak time at this time. The peak time of the representative scene is extracted and connected by lines.

[0140] The peak moments of the representative scenes extracted by the embodiment of the present application are all consistent with the locations of the peak moments of the original scenes, and the representative scenes of most clusters can coincide with the high-probability peak moments of the original scenes, indicating that the clustering result of the representative scene extraction model of the embodiment of the present application is optimal under this evaluation indicator.

[0141] For example, in order to verify that the representative photovoltaic power generation scene maintains the shape similarity characteristics of the original scene, the morphological trend similarity distance between the photovoltaic power generation data in the cluster and the representative scene of this type is calculated using the method described above, and the maximum similarity distance between the photovoltaic power generation data in each cluster and the representative scene of the cluster is compared. The smaller the value, the more similar the photovoltaic power generation data is to the representative scene of the cluster. The results of each comparison technique are listed in Table 2.

[0142] Table 2

[0143]

[0144] By comparing the morphological similarity distances in each cluster, compared with other technologies, most of the similarity distances calculated by the representative scene extraction model of this application are minimum values, indicating that the clustering result of the representative scene extraction model of this article is optimal under this evaluation indicator; by comparing the maximum values ​​of these 13 clusters, the maximum value of the morphological similarity distance calculated by the representative scene extraction model of this article is the smallest among these compared models, indicating that the clustering result of the representative scene extraction model of this article is optimal under this evaluation indicator.

[0145] According to the data processing method of the photovoltaic power station proposed in the embodiment of the present application, by performing multi-dimensional data clustering on the time series data, a multi-dimensional similarity measurement of the photovoltaic power station is obtained, which can more comprehensively reflect the actual characteristics of photovoltaic power generation, effectively extract representative photovoltaic scenes, reduce the amount of data, and improve the efficiency of data analysis and processing.

[0146] Next, a data processing device for a photovoltaic power station according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0147] Figure 7Schematic diagram of a data processing device for a photovoltaic power station according to an embodiment of the present application.

[0148] like Figure 7 As shown, the data processing device 10 of the photovoltaic power station includes: an acquisition module 100 , a clustering module 200 and a generation module 300 .

[0149] Among them, the acquisition module 100 is used to acquire the time series data of the photovoltaic power station; the clustering module 200 is used to perform multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station, wherein the multi-dimensionality is multiple dimensions of power, energy and volatility of the photovoltaic power station; the generation module 300 is used to generate a target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station.

[0150] In one embodiment of the present application, the clustering module 200 is further used to pre-cluster the time series data to obtain a target number of clusters; the data of the target number of clusters is input into the clustering model, and the clustering model outputs a multidimensional similarity measure of the photovoltaic power station, wherein the clustering model calculates the similarity measure, the similarity of the amount of electricity, the length of the mutual correlation sequence and the similarity of the fluctuation position of any two power time series in the data of the target number of clusters, and generates a multidimensional similarity measure based on the similarity measure, the similarity of the amount of electricity, the length of the mutual correlation sequence and the similarity of the fluctuation position.

[0151] In one embodiment of the present application, the clustering model is:

[0152] LESD(x i ,x j ,W)=w1L(x i ,x j )+w2E(x i ,x j )+w3S(x i ,x j )+w4δ(x i ,x j )

[0153] Among them, w1, w2, w3 and w4 represent weight coefficients, forming the vector W∈R 4×1 , x i and x j Represents a power time series of length L, x it , x jt Respectively represent the power time series x at time t i and x j Features;

[0154] E(x i ,xj ) represents x i and x j The similarity of the electric quantity;

[0155] CC ω (x i ,x j )=R ω-L (x i ,x j ),ω∈{1,2,K,2L-1},

[0156] CC w (x i ,x j ) represents x i and x j The cross-correlation sequence of s represents the power time series x j The moving step length, s∈[-L,L]; w is the cross-correlation sequence CC w (x i ,x j ) position subscript;

[0157] S(x i ,x j )=1-Max(NCC c (x i ,x j )),S(x i ,x j ) represents x i With x j The morphological trend similarity distance has a range of [0,2];

[0158] Represents the sequence x j Slide to and x i The displacement when the morphological trend is most similar, δ(x i ,x j ) indicates the similarity of the fluctuations of the two power time series.

[0159] In one embodiment of the present application, the data processing device 10 of the photovoltaic power station also includes: a determination module, which is used to obtain the neighborhood probability of the power time series and other sequences before inputting the data of the target number of clusters into the clustering model; and determine the neighborhood sequence of the power time series according to the neighborhood probability.

[0160] In one embodiment of the present application, the generation module 300 is further used to extract a morphologically representative scene from the multidimensional similarity measurement of the photovoltaic power station; the morphologically representative scene is amplified to obtain a target photovoltaic scene of the photovoltaic power station, wherein the target photovoltaic scene is:

[0161]

[0162] Among them, N g is the number of power time series of the gth category; x g,i,t is the characteristic value of the i-th power time series in the g-th category at time t; μ g,t * is the characteristic value of the representative scene of the g-th type of morphology at time t. P g The original sequence set representing the cluster where the representative scene to be sought belongs.

[0163] In one embodiment of the present application, the data processing device 10 of the photovoltaic power station further includes: a processing module for standardizing the time series data before performing multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station.

[0164] In one embodiment of the present application, the data processing device 10 of the photovoltaic power station also includes: an evaluation module, which is used to obtain multiple indicator data of random power flow, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period of photovoltaic grid connection after generating the target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station; based on the multiple indicator data of random power flow, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period of photovoltaic grid connection, evaluate the accuracy level of the target photovoltaic scene.

[0165] It should be noted that the above explanation of the embodiment of the data processing method for a photovoltaic power station is also applicable to the data processing device for a photovoltaic power station of this embodiment, and will not be repeated here.

[0166] According to the data processing device of the photovoltaic power station proposed in the embodiment of the present application, by performing multi-dimensional data clustering on time series data, a multi-dimensional similarity measurement of the photovoltaic power station is obtained, which can more comprehensively reflect the actual characteristics of photovoltaic power generation, effectively extract representative photovoltaic scenes, reduce the amount of data, and improve the efficiency of data analysis and processing.

[0167] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0168] A memory 801 , a processor 802 , and a computer program stored in the memory 801 and executable on the processor 802 .

[0169] When the processor 802 executes the program, the data processing method for the photovoltaic power station provided in the above embodiment is implemented.

[0170] Furthermore, the electronic device further comprises:

[0171] The communication interface 803 is used for communication between the memory 801 and the processor 802 .

[0172] The memory 801 is used to store computer programs that can be executed on the processor 802 .

[0173] The memory 801 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0174] If the memory 801, the processor 802 and the communication interface 803 are implemented independently, the communication interface 803, the memory 801 and the processor 802 can be connected to each other through a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0175] Optionally, in a specific implementation, if the memory 801, the processor 802 and the communication interface 803 are integrated on a chip, the memory 801, the processor 802 and the communication interface 803 can communicate with each other through an internal interface.

[0176] The processor 802 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0177] The embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the data processing method of the photovoltaic power station as described above is implemented.

[0178] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0179] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0180] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0181] It should be understood that the various parts of the present application can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, the steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array, a field programmable gate array, etc.

[0182] A person of ordinary skill in the art may understand that all or part of the steps carried by the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the above-mentioned program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.

[0183] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A data processing method for a photovoltaic power station, characterized in that: The following steps are involved: Obtain time series data of photovoltaic power plants; Performing multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measure of the photovoltaic power station, wherein the multi-dimensionality is a plurality of power dimensions, energy dimensions, and volatility dimensions of the photovoltaic power station; A target photovoltaic scene of the photovoltaic power station is generated according to the multi-dimensional similarity measurement of the photovoltaic power station.

2. The data processing method of a photovoltaic power station according to claim 1, characterized in that: The multi-dimensional data clustering of the time series data to obtain the multi-dimensional similarity measurement of the photovoltaic power station includes: Pre-clustering the time series data to obtain a target number of clusters; The data of the target number of clusters is input into a clustering model, and the clustering model outputs a multidimensional similarity metric of the photovoltaic power station, wherein the clustering model calculates the similarity metric, the similarity of the amount of electricity, the length of the mutual correlation sequence and the similarity of the fluctuation position of any two power time series in the data of the target number of clusters, and generates a multidimensional similarity metric according to the similarity metric, the similarity of the amount of electricity, the length of the mutual correlation sequence and the similarity of the fluctuation position.

3. The data processing method of a photovoltaic power station according to claim 2, characterized in that: The clustering model is: LESD(x i ,x j ,W)=w1L(x i ,x j )+w2E(x i ,x j )+w3S(x i ,x j )+w4δ(x i ,x j ) Among them, w1, w2, w3 and w4 represent weight coefficients, forming the vector W∈R 4×1 , x i and x j Represents a power time series of length L, x it , x jt Respectively represent the power time series x at time t i and x j Features; E(x i ,x j ) represents x i and x j The similarity of the electric quantity; CC ω (x i ,x j )=R ω-L (x i ,x j ),ω∈{1,2,K,2L-1}, CC w (x i ,x j ) represents x i and x j The cross-correlation sequence of s represents the power time series x j The moving step length, s∈[-L,L]; w is the cross-correlation sequence CC w (x i ,x j ) position subscript; S(x i ,x j )=1-Max(NCC c (x i ,x j )),S(x i ,x j ) represents x i With x j The morphological trend similarity distance has a range of [0,2]; Represents the sequence x j Slide to and x i The displacement when the morphological trend is most similar, δ(x i ,x j ) indicates the similarity of the fluctuations of the two power time series.

4. The data processing method of a photovoltaic power station according to claim 2, characterized in that: Before inputting the data of the target number of clusters into the clustering model, the method further comprises: Get the neighborhood probabilities of the power time series and other series; A neighborhood sequence of the power time series is determined according to the neighborhood probability.

5. The data processing method of a photovoltaic power station according to claim 1, characterized in that: The step of generating a target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station comprises: extracting morphologically representative scenes from a multidimensional similarity measure of the photovoltaic power station; The representative morphological scene is magnified to obtain a target photovoltaic scene of the photovoltaic power station, wherein the target photovoltaic scene is: Among them, N g is the number of power time series of the gth category; x g,i,t is the characteristic value of the i-th power time series in the g-th category at time t, is the representative scene of the g-th type of morphology, μ g,t * is the characteristic value of the representative scene of the g-th type of morphology at time t, P g The original sequence set representing the cluster where the representative scene to be sought belongs.

6. The data processing method of a photovoltaic power station according to claim 1, characterized in that: Before performing multi-dimensional data clustering on the time series data to obtain the multi-dimensional similarity measurement of the photovoltaic power station, the method further includes: The time series data is standardized.

7. The data processing method of a photovoltaic power station according to claim 1, characterized in that: After generating the target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station, the method further includes: Obtain multiple indicator data of random power flow, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value of photovoltaic power generation peak period of photovoltaic grid connection; Based on multiple indicator data such as the random power flow of the photovoltaic grid-connected system, annual power generation, similarity of photovoltaic daily power generation curve shape and frequency value during photovoltaic power generation peak period, the accuracy level of the target photovoltaic scenario is evaluated.

8. A data processing device for a photovoltaic power station, characterized in that: include: An acquisition module is used to obtain time series data of photovoltaic power plants; A clustering module, used for performing multi-dimensional data clustering on the time series data to obtain a multi-dimensional similarity measurement of the photovoltaic power station, wherein the multi-dimensionality is a plurality of power dimensions, energy dimensions and volatility dimensions of the photovoltaic power station; A generating module is used to generate a target photovoltaic scene of the photovoltaic power station according to the multi-dimensional similarity measurement of the photovoltaic power station.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the data processing method for a photovoltaic power station according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed, the data processing method for a photovoltaic power station according to any one of claims 1 to 7 is implemented.

Images