Distributed photovoltaic power anomaly discrimination method and related apparatus

Abstract

Embodiments of the present application relate to the field of data analysis, and provide a distributed photovoltaic power abnormality discrimination method and related device, the method comprising: performing normalization processing on collected photovoltaic power data corresponding illumination intensity data to obtain a first illumination intensity data set; performing clustering processing on photovoltaic power data in a photovoltaic power data set according to the first illumination intensity data set to obtain K first photovoltaic power data sets; performing elimination processing on abnormal data in the first photovoltaic power data sets to obtain a second photovoltaic power data set; calculating the similarity between each second photovoltaic power data in the second photovoltaic power data set to obtain a similarity set; eliminating abnormal data in the second photovoltaic power data set according to the profile coefficient corresponding to each second photovoltaic power data calculated by the similarity set to obtain a fourth photovoltaic power data set, thereby improving the accuracy of photovoltaic power data in abnormality discrimination.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a distributed photovoltaic power abnormality identification method and related device. BACKGROUND

[0002] When the output power of photovoltaic power supply has the characteristics of intermittence and fluctuation, the access of distributed photovoltaic power generation increases the difficulty of load prediction, thereby affecting the scheduling and operation plan of the power grid. In addition, if the output power of photovoltaic power generation changes in the opposite direction of the system load, it may further expand the voltage fluctuation of the system, and in severe cases, it may cause system voltage flicker, which threatens the safe and stable operation of the power grid. On the other hand, when the photovoltaic power generation network is connected to the distribution network, it can increase the reactive power output of the distribution network system and reduce the transmission power of the line. If the capacity of the connected photovoltaic power supply is too large, it may cause the voltage of some nodes to exceed the standard, resulting in voltage deviation, voltage fluctuation, and voltage flicker, and other voltage quality problems. In the existing scheme, when the photovoltaic power is abnormally identified, it is usually directly identified, so that the accuracy of the abnormal identification is low. SUMMARY

[0003] The embodiments of the present application provide a distributed photovoltaic power abnormality identification method and related device, which can improve the accuracy of abnormal identification of photovoltaic power.

[0004] The first aspect of the embodiments of the present application provides a distributed photovoltaic power abnormality identification method, which comprises:

[0005] The light intensity data corresponding to the photovoltaic power data in the collected photovoltaic power data set is normalized to obtain a first light intensity data set;

[0006] The photovoltaic power data in the photovoltaic power data set is clustered according to the first light intensity data set to obtain K first photovoltaic power data sets;

[0007] The abnormal data in the corresponding first photovoltaic power data set is removed according to the allowed threshold corresponding to the K first photovoltaic power data sets to obtain a second photovoltaic power data set;

[0008] The similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set;

[0009] The profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set is calculated according to the similarity set to obtain a profile coefficient set;

[0010] The abnormal data in the second photovoltaic power data set is removed according to the profile coefficient set to obtain a fourth photovoltaic power data set.

[0011] In the example, the corresponding illumination intensity data of the photovoltaic power data in the collected photovoltaic power data set is normalized to obtain a first illumination intensity data set; the photovoltaic power data in the photovoltaic power data set is clustered according to the first illumination intensity data set to obtain K first photovoltaic power data sets; the abnormal data in the corresponding first photovoltaic power data set is removed according to the corresponding allowed threshold of the K first photovoltaic power data sets to obtain a second photovoltaic power data set; the similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set; the profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set is calculated according to the similarity set to obtain a profile coefficient set; the abnormal data in the second photovoltaic power data set is removed according to the profile coefficient set to obtain a fourth photovoltaic power data set. Therefore, the abnormal data in the photovoltaic power data set can be preliminarily removed to obtain a second photovoltaic power data set, and the similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set, the profile coefficient corresponding to each second photovoltaic power data is calculated according to the similarity set to obtain a profile coefficient set, and the second photovoltaic power data in the second photovoltaic power data set is distinguished according to the profile coefficient set, thereby improving the accuracy of distinguishing abnormal photovoltaic power data.

[0012] In one possible implementation, a method for normalizing the corresponding illumination intensity data of the photovoltaic power data in the collected photovoltaic power data set to obtain a first illumination intensity data set, comprising:

[0013] The corresponding illumination intensity data of the photovoltaic power data in the collected photovoltaic power data set is normalized by the following formula to obtain a first illumination intensity data set:

[0014]

[0015] wherein, S represents the second illumination intensity data in the second illumination intensity data set; S t S represents the first illumination intensity data in the first illumination intensity data set; S max S represents the maximum illumination intensity data in the first illumination intensity data set; S min S represents the minimum illumination intensity data in the first illumination intensity data set.

[0016] In a possible implementation, a method for clustering photovoltaic power data in a photovoltaic power data set according to a first light intensity data set to obtain K first photovoltaic power data sets, comprises the following steps:

[0017] A first number of cluster centers corresponding to the photovoltaic power data in the photovoltaic power data set is determined by using the elbow method, to obtain the first number.

[0018] The photovoltaic power data in the photovoltaic power data set is clustered into K clusters by using the K-means clustering method according to the first number, to obtain K second photovoltaic power data sets.

[0019] In a possible implementation, a method for removing abnormal data in the K first photovoltaic power data sets according to the allowed threshold corresponding to the K first photovoltaic power data sets respectively to obtain a second photovoltaic power data set, comprises the following steps:

[0020] A first photovoltaic power matrix is constructed according to a target first photovoltaic power data set, the target first photovoltaic power data set being any one of the K first photovoltaic power data sets.

[0021] A photovoltaic power average value matrix is obtained by calculating the photovoltaic power average value corresponding to each time period according to the first photovoltaic power matrix.

[0022] A target photovoltaic power average value is obtained according to the photovoltaic power average value matrix.

[0023] A reference first photovoltaic data set is obtained by removing the target first photovoltaic data in the target first photovoltaic data set, the absolute value of the difference between which and the target photovoltaic power average value is greater than the allowed threshold.

[0024] The steps of constructing a first photovoltaic power matrix according to a target first photovoltaic power data set, and removing the target first photovoltaic data in the target first photovoltaic data set, the absolute value of the difference between which and the target photovoltaic power average value is greater than the allowed threshold, to obtain a reference first photovoltaic data set, are repeatedly performed until the reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets respectively are obtained.

[0025] The union of the reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets respectively is determined as a second photovoltaic power data set.

[0026] In a possible implementation, a method for calculating the similarity between each second photovoltaic power data in the second photovoltaic power data set to obtain a similarity set, comprises the following steps:

[0027] constructing a photovoltaic power and illumination intensity time series matrix corresponding to the second photovoltaic power data set;

[0028] calculating, according to the photovoltaic power and illumination intensity time series matrix, a Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set, to obtain a Euclidean distance set;

[0029] determining a dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set, to obtain a dynamic time warping distance set;

[0030] constructing a similarity function according to the Euclidean distance set and the dynamic time warping distance set;

[0031] calculating, according to the Euclidean distance set and the dynamic time warping distance set, a similarity between each second photovoltaic power data in the second photovoltaic power data set by the similarity function, to obtain a similarity set.

[0032] A second aspect of the embodiment of the application provides a distributed photovoltaic power anomaly identification device, which comprises:

[0033] a processing unit configured to normalize illumination intensity data corresponding to photovoltaic power data in a collected photovoltaic power data set, to obtain a first illumination intensity data set;

[0034] a clustering unit configured to perform clustering processing on the photovoltaic power data in the photovoltaic power data set according to the first illumination intensity data set, to obtain K first photovoltaic power data sets;

[0035] a first elimination unit configured to perform elimination processing on abnormal data in a corresponding first photovoltaic power data set according to an allowed threshold corresponding to the K first photovoltaic power data sets, to obtain a second photovoltaic power data set;

[0036] a first calculation unit configured to calculate a similarity between each second photovoltaic power data in the second photovoltaic power data set, to obtain a similarity set;

[0037] a second calculation unit configured to calculate a profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set according to the similarity set, to obtain a profile coefficient set;

[0038] a second elimination unit configured to eliminate abnormal data in the second photovoltaic power data set according to the profile coefficient set, to obtain a fourth photovoltaic power data set.

[0039] In one possible implementation, the first elimination unit is specifically configured to:

[0040] constructing a photovoltaic power data matrix according to the target first photovoltaic power data set, to obtain a first photovoltaic power matrix, the target first photovoltaic power data set being any one of K first photovoltaic power data sets;

[0041] calculating a photovoltaic power average value corresponding to each time period according to the first photovoltaic power matrix, to obtain a photovoltaic power average value matrix;

[0042] obtaining a target photovoltaic power average value according to the photovoltaic power average value matrix;

[0043] eliminating target first photovoltaic data in the target first photovoltaic data set, which has an absolute value greater than an allowable threshold corresponding to a difference between the target photovoltaic power average value, to obtain a reference first photovoltaic data set;

[0044] repeating the steps of constructing a photovoltaic power data matrix according to the target first photovoltaic power data set, to obtain a first photovoltaic power matrix, to the step of eliminating target first photovoltaic data in the target first photovoltaic data set, which has an absolute value greater than an allowable threshold corresponding to a difference between the target photovoltaic power average value, to obtain a reference first photovoltaic data set, until K reference first photovoltaic data sets corresponding to K first photovoltaic power data sets are obtained;

[0045] determining a union of the K reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets as a second photovoltaic power data set.

[0046] In one possible implementation, the first calculation unit is specifically configured to:

[0047] constructing a photovoltaic power and irradiance time series matrix corresponding to the second photovoltaic power data set;

[0048] calculating a Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set according to the photovoltaic power and irradiance time series matrix, to obtain a Euclidean distance set;

[0049] determining a dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set, to obtain a dynamic time warping distance set;

[0050] constructing a similarity function according to the Euclidean distance set and the dynamic time warping distance set;

[0051] calculating a similarity between each second photovoltaic power data in the second photovoltaic power data set through the similarity function according to the Euclidean distance set and the dynamic time warping distance set, to obtain a similarity set.

[0052] In one possible implementation, the first elimination unit is specifically configured to:

[0053] constructing a photovoltaic power data matrix according to the target first photovoltaic power data set, to obtain a first photovoltaic power matrix, the target first photovoltaic power data set being any one of the K first photovoltaic power data sets;

[0054] calculating a photovoltaic power average value corresponding to each time period according to the first photovoltaic power matrix, to obtain a photovoltaic power average value matrix;

[0055] obtaining a target photovoltaic power average value according to the photovoltaic power average value matrix;

[0056] eliminating target first photovoltaic data in the target first photovoltaic data set, which has an absolute value greater than an allowable threshold corresponding to a difference between the target photovoltaic power average value, to obtain a reference first photovoltaic data set;

[0057] repeating the steps of constructing a photovoltaic power data matrix according to the target first photovoltaic power data set, to obtain a first photovoltaic power matrix, to the step of eliminating target first photovoltaic data in the target first photovoltaic data set, which has an absolute value greater than an allowable threshold corresponding to a difference between the target photovoltaic power average value, to obtain a reference first photovoltaic data set, until K reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets are obtained;

[0058] determining a union of the K reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets as a second photovoltaic power data set.

[0059] In one possible implementation, the first calculation unit is specifically configured to:

[0060] constructing a photovoltaic power and irradiance time series matrix corresponding to the second photovoltaic power data set;

[0061] calculating a Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set according to the photovoltaic power and irradiance time series matrix, to obtain a Euclidean distance set;

[0062] determining a dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set, to obtain a dynamic time warping distance set;

[0063] constructing a similarity function according to the Euclidean distance set and the dynamic time warping distance set;

[0064] calculating a similarity between each second photovoltaic power data in the second photovoltaic power data set by the similarity function according to the Euclidean distance set and the dynamic time warping distance set, to obtain a similarity set.

[0065] A third aspect of the embodiments of the present application provides a terminal, comprising a processor, an input device, an output device and a memory, which are connected with each other, wherein the memory is configured to store a computer program, the computer program comprises program instructions, and the processor is configured to invoke the program instructions to execute the step instructions as in the first aspect of the embodiments of the present application.

[0066] A fourth aspect of the embodiments of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program for electronic data exchange, and the computer program causes a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application.

[0067] A fifth aspect of the embodiments of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application. The computer program product can be a software installation package. BRIEF DESCRIPTION OF DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0069] Figure 1 A flowchart of a distributed photovoltaic power abnormality discrimination method is provided for the embodiments of the present application;

[0070] Figure 2 A structural diagram of a terminal is provided for the embodiments of the present application;

[0071] Figure 3 A structural diagram of a distributed photovoltaic power abnormality discrimination device is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0072] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0073] The terms "first", "second", and the like in the description and in the claims of the present application and the above drawings are used to distinguish different objects, and are not used to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device including a series of steps or units is not limited to the listed steps or units, but can optionally further include steps or units not listed, or can optionally further include other steps or units inherent to the process, method, product, or device.

[0074] In the present application, the phrase "embodiment" means that the specific features, structures or characteristics described in conjunction with the embodiment can be included in at least one embodiment of the present application. The appearance of this phrase at various places in the specification does not necessarily refer to the same embodiment, nor is it independent or alternative to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in the present application can be combined with other embodiments.

[0075] In order to better understand the distributed photovoltaic power anomaly identification method provided by the embodiments of the present application, the distributed photovoltaic power anomaly identification method in the prior art will be briefly introduced first. In the prior art, the photovoltaic power data corresponding to the light intensity data of each time period is usually predicted according to the light intensity data of each time period combined with a preset algorithm to obtain a photovoltaic power prediction value; then the photovoltaic power prediction value is compared with the corresponding photovoltaic power data collected to determine whether the difference between the photovoltaic power data and the corresponding photovoltaic power prediction value is less than a preset threshold; if the difference between the photovoltaic power data and the corresponding photovoltaic power prediction value is less than or equal to the preset threshold, it means that the collected photovoltaic power data is normal data; otherwise, it means that the collected photovoltaic power data is abnormal data. However, the photovoltaic power data is affected by many factors, resulting in inaccurate photovoltaic power prediction values, thereby reducing the accuracy of photovoltaic power data anomaly identification.

[0076] To solve the above technical problems, the embodiments of the present application provide a distributed photovoltaic power anomaly identification method, which can calculate the similarity between each photovoltaic power data collected to obtain a similarity set, calculate the contour coefficient corresponding to each photovoltaic power data according to the similarity set to obtain a contour coefficient set, and identify the photovoltaic power data according to the contour coefficient set, thereby improving the efficiency of photovoltaic power data anomaly identification.

[0077] Please refer to Figure 1 , Figure 1 A flowchart of a distributed photovoltaic power anomaly identification method is provided for the embodiments of the present application. As shown in Figure 1 , the method comprises:

[0078] 101. Normalize the light intensity data corresponding to the photovoltaic power data in the collected photovoltaic power data set to obtain the first light intensity data set.

[0079] Specifically, the irradiance data corresponding to the photovoltaic power data in the photovoltaic power dataset can be normalized using a pre-set algorithm in the server. This can be achieved by normalizing the irradiance data corresponding to the photovoltaic power data in the collected photovoltaic power dataset using the method shown in the following formula, thus obtaining the first irradiance dataset:

[0080]

[0081] in, S represents the second illumination intensity data in the second illumination intensity data set; t S represents the first illuminance data in the first illuminance data set; max S represents the maximum illuminance data in the first illuminance data set; min This represents the minimum light intensity data in the first set of light intensity data.

[0082] 102. Based on the first light intensity data set, cluster the photovoltaic power data in the photovoltaic power data set to obtain K first photovoltaic power data sets.

[0083] Specifically, the number of clusters corresponding to the photovoltaic power data in the photovoltaic power data set can be calculated using the first set of light intensity data, and the photovoltaic power data can be divided into K first photovoltaic power data sets based on the number of clusters.

[0084] 103. Based on the allowable thresholds corresponding to the K first photovoltaic power data sets, the abnormal data in the corresponding first photovoltaic power data sets are removed to obtain the second photovoltaic power data set.

[0085] Specifically, the target photovoltaic power average value can be calculated by constructing a photovoltaic power matrix corresponding to each first photovoltaic power data set. Abnormal data in each first photovoltaic power data set can be removed based on the target photovoltaic power average value and the corresponding allowable threshold, resulting in K reference first photovoltaic power data sets. The K reference first photovoltaic power data sets can then be merged to obtain a second photovoltaic power data set.

[0086] 104. Calculate the similarity between each second photovoltaic power data in the second photovoltaic power data set to obtain a similarity set.

[0087] The Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set can be calculated by constructing a light intensity time sequence matrix corresponding to the second photovoltaic power data set, to obtain a Euclidean distance set, a dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set is obtained, a dynamic time warping distance set is obtained, a similarity function is constructed according to the Euclidean distance set and the dynamic time warping distance set, and the similarity between each second photovoltaic power data in the second photovoltaic power data set is determined according to the similarity function and the first light intensity data corresponding to each second photovoltaic power data in the second photovoltaic power data set, to obtain a similarity set.

[0088] 105. Calculate the profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set according to the similarity set, to obtain a profile coefficient set.

[0089] The second photovoltaic power data in the second photovoltaic power data set can be classified according to the similarity set, and then the first average value of the Euclidean distance between each second photovoltaic power data and other second photovoltaic data belonging to the same category is obtained, and the second average value of the Euclidean distance between each second photovoltaic power data and other second photovoltaic data belonging to the most similar category is obtained. The profile coefficient corresponding to each second photovoltaic power data is calculated according to the first average value and the second average value corresponding to each second photovoltaic power data, to obtain a profile coefficient set.

[0090] 106. Remove abnormal data in the second photovoltaic power data set according to the profile coefficient set, to obtain a fourth photovoltaic power data set.

[0091] The profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set can be determined by judging whether the profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set is close to 1. If the profile coefficient corresponding to the second photovoltaic power data is close to 1, it means that the second photovoltaic power data is normal data. Otherwise, it means that the second photovoltaic power data is abnormal data, and the second photovoltaic power data is removed. Until all abnormal data in the second photovoltaic power data set is removed, a fourth photovoltaic power data set is obtained.

[0092] In the example, the corresponding illumination intensity data of the photovoltaic power data in the collected photovoltaic power data set is normalized to obtain a first illumination intensity data set; the photovoltaic power data in the photovoltaic power data set is clustered according to the first illumination intensity data set to obtain K first photovoltaic power data sets; the abnormal data in the corresponding first photovoltaic power data set is removed according to the corresponding allowed threshold of the K first photovoltaic power data sets to obtain a second photovoltaic power data set; the similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set; the profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set is calculated according to the similarity set to obtain a profile coefficient set; the abnormal data in the second photovoltaic power data set is removed according to the profile coefficient set to obtain a fourth photovoltaic power data set, so that the abnormal data in the photovoltaic power data set can be preliminarily removed to obtain a second photovoltaic power data set, and the similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set, the profile coefficient corresponding to each second photovoltaic power data is calculated according to the similarity set to obtain a profile coefficient set, and the second photovoltaic power data in the second photovoltaic power data set is distinguished according to the profile coefficient set, thereby improving the accuracy of distinguishing the abnormal photovoltaic power data.

[0093] In one possible implementation, a method for normalizing the corresponding illumination intensity data of the photovoltaic power data in the collected photovoltaic power data set to obtain a first illumination intensity data set, comprising:

[0094] The corresponding illumination intensity data of the photovoltaic power data in the collected photovoltaic power data set can be normalized by the following formula to obtain a first illumination intensity data set:

[0095]

[0096] Wherein, S represents the second illumination intensity data in the second illumination intensity data set; S t S represents the first illumination intensity data in the first illumination intensity data set; S max S represents the maximum illumination intensity data in the first illumination intensity data set; S min S represents the minimum illumination intensity data in the first illumination intensity data set.

[0097] In the example, the corresponding illumination intensity data of the photovoltaic power data in the photovoltaic power data set is normalized to improve the stability of the data in subsequent data processing, thereby improving the accuracy of the photovoltaic power data in the abnormal identification.

[0098] In a possible implementation, a method for clustering photovoltaic power data in a photovoltaic power data set according to a first light intensity data set to obtain K first photovoltaic power data sets, comprising:

[0099] B1, determining the number of cluster centers corresponding to the photovoltaic power data in the photovoltaic power data set by using the elbow method, to obtain a first number;

[0100] B2, clustering the photovoltaic power data in the photovoltaic power data set into K clusters according to the first number by using the K-means clustering method, to obtain K second photovoltaic power data sets.

[0101] Wherein, the cluster center corresponding to the photovoltaic power data in the photovoltaic power data set can be determined by the elbow method as shown in the following formula:

[0102]

[0103] In the formula, S sE represents the first light intensity information corresponding to the cluster center; K represents the number of clusters; C i represents the i-th second photovoltaic power data set; p i represents the i-th second photovoltaic power data in C i ; and m i represents the cluster center of C i . After clustering all the cluster centers, the first number is obtained.

[0104] In a possible implementation, a method for removing abnormal data in K first photovoltaic power data sets according to the allowed threshold corresponding to the K first photovoltaic power data sets respectively, to obtain a second photovoltaic power data set, comprising:

[0105] C1, constructing a photovoltaic power data matrix according to a target first photovoltaic power data set to obtain a first photovoltaic power matrix, the target first photovoltaic power data set being any one of the K first photovoltaic power data sets;

[0106] C2, calculating the photovoltaic power average value corresponding to each time period according to the first photovoltaic power matrix to obtain a photovoltaic power average value matrix;

[0107] C3, obtaining a target photovoltaic power average value according to the photovoltaic power average value matrix;

[0108] C4, removing the target first photovoltaic data whose absolute value corresponding to the difference between the target first photovoltaic data and the target photovoltaic power average value is greater than the allowed threshold, to obtain a reference first photovoltaic data set.

[0109] C5, repeatedly performing the above-mentioned construction of the photovoltaic power data matrix according to the target first photovoltaic power data set to obtain a first photovoltaic power matrix until the step of removing the target first photovoltaic data in the target first photovoltaic data set corresponding to the absolute value greater than the allowed threshold value between the target photovoltaic power average value to obtain the reference first photovoltaic data set, until K first photovoltaic power data sets respectively corresponding to the reference first photovoltaic data set are obtained;

[0110] C6, the union of the reference first photovoltaic data set corresponding to the K first photovoltaic power data set is determined as a second photovoltaic power data set;

[0111] The first photovoltaic power matrix can be:

[0112]

[0113] In the formula, P pv,t represents the first photovoltaic power matrix at t period; P m,t,n represents the mth first photovoltaic power data in the target first photovoltaic power data set at t period of the nth day.

[0114] The photovoltaic power average value matrix can be:

[0115]

[0116] In the formula, represents the average value corresponding to all first photovoltaic power data at t period of the nth day; n represents the number of days. The photovoltaic power average value can be calculated by using a general photovoltaic power average value calculation method.

[0117] Before obtaining the target photovoltaic power average value, the optimal target photovoltaic power average value threshold factor also needs to be obtained. Specifically, the optimal target photovoltaic power average value threshold factor can be obtained by the method shown in the following formula:

[0118]

[0119] In the formula, represents the optimal target photovoltaic power average value threshold factor at t period; represents the reference first photovoltaic power data in the reference first photovoltaic power data set at t period; represents the correlation between the illumination intensity and the photovoltaic power of the mth reference first photovoltaic power data in the reference first photovoltaic power data set at t period; represents the maximum correlation in the correlation matrix; represents the threshold factor removal rate; η max represents the upper limit of the threshold factor removal rate.

[0120] After obtaining the optimal target photovoltaic power average value threshold factor, the target photovoltaic power average value can be obtained by calculating the product of the optimal target photovoltaic power average value threshold factor and the maximum value in the photovoltaic power average value matrix. Specifically, the product of the optimal target photovoltaic power average value threshold factor and the maximum value in the photovoltaic power average value matrix can be calculated by the method shown in the following formula to obtain the target photovoltaic power average value:

[0121]

[0122] In the formula, P t，max represents the maximum value in the first photovoltaic power matrix; represents the first photovoltaic power data in the first photovoltaic power matrix; P t represents the target photovoltaic power average value; λ t represents the optimal target photovoltaic power average value threshold factor.

[0123] After obtaining the target photovoltaic power average value, the absolute value of the difference between each target first photovoltaic power data in the target first photovoltaic power data set and the target photovoltaic power average value can be calculated. Then, the size relationship between the absolute value of the difference between each target first photovoltaic power data and the target photovoltaic power average value and the allowable threshold is judged. If the absolute value of the difference is less than the allowable threshold, it means that the target first photovoltaic power data is normal data. If the absolute value of the difference is greater than the allowable threshold, it means that the target first photovoltaic power data is abnormal data. The abnormal data in the target first photovoltaic power data set is removed to obtain the reference first photovoltaic power data set. The specific reference first photovoltaic data set is:

[0124]

[0125] In the formula, Ψ represents the reference first photovoltaic data set; P j,t,i represents the i-th target first photovoltaic power data at t time period on the i-th day in the target first photovoltaic power data set; represents the target photovoltaic power average value; δ pv represents the allowable threshold.

[0126] In one possible implementation, a method for calculating the similarity between each second photovoltaic power data in the second photovoltaic power data set to obtain a similarity set includes:

[0127] D1, constructing a photovoltaic power and light intensity time sequence matrix corresponding to the second photovoltaic power data set;

[0128] D2, calculate the Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set according to the photovoltaic power and light intensity time series matrix, to obtain a Euclidean distance set;

[0129] D3, determine the dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set, to obtain a dynamic time warping distance set.

[0130] D4, construct a similarity function according to the Euclidean distance set and the dynamic time warping distance set

[0131] Wherein, the risk assessment result includes risk level, the corresponding risk level can be extracted from the risk assessment result.

[0132] D5, calculate the similarity between each second photovoltaic power data in the second photovoltaic power data set according to the Euclidean distance set and the dynamic time warping distance set through the similarity function, to obtain a similarity set.

[0133] Wherein, the photovoltaic power and light intensity time series matrix corresponding to the second photovoltaic power data set can be represented as:

[0134]

[0135] In the formula, D represents the photovoltaic power and light intensity time series matrix corresponding to the second photovoltaic power data set, d Pi,Sj Indicates the corresponding relationship of the elements in the two time series, i,j∈[1,2,…,n].

[0136] The Euclidean distance between the second photovoltaic power data can be calculated according to the photovoltaic power and light intensity time series matrix by the following formula:

[0137]

[0138] In the formula, d(i,j) represents the Euclidean distance between the ith second photovoltaic power data and the jth photovoltaic power data; P i Indicates the ith second photovoltaic power data in the time series matrix; S j Indicates the first light intensity data corresponding to the jth second photovoltaic power data in the time series matrix.

[0139] After obtaining the set of Euclidean distances, in order to make the similarity between each second photovoltaic power data obtained by subsequent calculation more reliable, it is also necessary to time regularize the Euclidean distances in the set of Euclidean distances to obtain a set of dynamic time warping distances. The dynamic time warping distance corresponding to each Euclidean distance in the set of Euclidean distances can be determined by a recursive method to obtain the set of dynamic time warping distances. Specifically, the dynamic time warping distance corresponding to each Euclidean distance in the set of Euclidean distances can be determined by the following formula:

[0140]

[0141] In the formula, D TW (i,j) represents the dynamic time warping distance corresponding to the Euclidean distance; d(i,j) represents the Euclidean distance; D TW (i-1,j) represents the dynamic time warping distance corresponding to the element in the i-1th column and the jth row of the photovoltaic power and light intensity time series matrix; D TW (i,j-1) represents the dynamic time warping distance corresponding to the element in the ith column and the j-1th row of the photovoltaic power and light intensity time series matrix; D TW (i-1,j-1) represents the dynamic time warping distance corresponding to the element in the i-1th column and the j-1th row of the photovoltaic power and light intensity time series matrix.

[0142] After obtaining the set of Euclidean distances and the set of dynamic time warping distances, the similarity function can be constructed by the preset similarity weight coefficient in the server, the set of Euclidean distances and the set of dynamic time warping distances, so that the server can input the Euclidean distance and the dynamic time warping distance between two second photovoltaic power data into the similarity function to calculate the similarity between the two second photovoltaic power data. The similarity function constructed according to the set of Euclidean distances and the set of dynamic time warping distances can be represented as:

[0143] C(i,j)=α1d(i,j)+α2D TW (i,j)+α3d(i Δ ,j Δ )+α4D TW (i Δ ,j Δ )

[0144] In the formula, C(i,j) represents the similarity corresponding to the element in the ith column and the jth row of the photovoltaic power and light intensity time series matrix; α1 represents the first similarity weight coefficient; d(i,j) represents the Euclidean distance corresponding to the element in the ith column and the jth row of the photovoltaic power and light intensity time series matrix; α2 represents the second similarity weight coefficient; D TW(i,j) represents the dynamic time warping distance corresponding to the element of the ith column and the jth row in the photovoltaic power and light intensity time series matrix; a3 represents the third similarity weight coefficient; di Δ ,j Δ ) represents the change rate of the Euclidean distance in the Euclidean distance set; a4 represents the fourth similarity weight coefficient; D TW (i Δ ,j Δ ) represents the change rate of the dynamic time warping distance in the dynamic time warping distance set; i Δ represents the change rate of the photovoltaic power data corresponding to the second photovoltaic power data in the photovoltaic power and light intensity time series matrix, which can be represented as where Δi represents the power parameter, which is a fixed value and can be determined by user input or system default; j Δ represents the change rate of the light intensity data corresponding to the second photovoltaic power data in the photovoltaic power and light intensity time series matrix, which can be represented as where Δj represents the power parameter, which is a fixed value and can be determined by user input or system default.

[0145] For the above embodiment, please refer to Figure 2 , Figure 2 A structure schematic diagram of a terminal provided by the embodiment of the present application is shown in Figure 2 , which includes a processor, an input device, an output device and a memory, and the processor, the input device, the output device and the memory are connected to each other, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to invoke the program instructions, and the above program includes instructions for executing the following steps;

[0146] The light intensity data corresponding to the photovoltaic power data in the collected photovoltaic power data set is normalized to obtain a first light intensity data set;

[0147] The photovoltaic power data in the photovoltaic power data set is clustered according to the first light intensity data set to obtain K first photovoltaic power data sets;

[0148] The abnormal data in the corresponding first photovoltaic power data set is removed according to the allowed threshold value corresponding to the K first photovoltaic power data sets respectively to obtain a second photovoltaic power data set;

[0149] The similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set;

[0150] According to the similarity set, a profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set is calculated, and a profile coefficient set is obtained;

[0151] According to the profile coefficient set, abnormal data in the second photovoltaic power data set is removed, and a fourth photovoltaic power data set is obtained.

[0152] The above mainly introduces the scheme of the embodiments of the present application from the perspective of the process of executing the method. It can be understood that the terminal includes a hardware structure and / or a software module corresponding to the execution of each function in order to realize the above functions. Those skilled in the art should easily realize that, in combination with the unit and algorithm steps of each example described in the embodiments provided herein, the present application can be realized in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the form of hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0153] The embodiments of the present application can divide the functional units of the terminal according to the above method examples. For example, each functional unit can be divided according to each function, or two or more functions can be integrated in one processing unit. The integrated unit can be realized in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division. Actual implementation can have another division manner.

[0154] Consistent with the above, please refer to Figure 3 , Figure 3 A structural schematic diagram of a distributed photovoltaic power abnormality discrimination device is provided for the embodiments of the present application. As shown in Figure 3 , the device includes:

[0155] The processing unit 301 is configured to normalize the light intensity data corresponding to the photovoltaic power data in the collected photovoltaic power data set, and obtain a first light intensity data set;

[0156] The clustering unit 302 is configured to perform clustering processing on the photovoltaic power data in the photovoltaic power data set according to the first light intensity data set, and obtain K first photovoltaic power data sets;

[0157] The first removing unit 303 is configured to remove abnormal data in the corresponding first photovoltaic power data set according to the allowed threshold corresponding to the K first photovoltaic power data sets, and obtain a second photovoltaic power data set;

[0158] The first computing unit 304 is configured to calculate the similarity between each second photovoltaic power data in the second photovoltaic power data set, to obtain a similarity set.

[0159] The second computing unit 305 is configured to calculate the profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set according to the similarity set, to obtain a profile coefficient set.

[0160] The second removing unit 306 is configured to remove abnormal data in the second photovoltaic power data set according to the profile coefficient set, to obtain a fourth photovoltaic power data set.

[0161] In one possible implementation, the first processing unit 301 is specifically configured to:

[0162] The photovoltaic power data corresponding to the photovoltaic power data in the collected photovoltaic power data set is normalized by the following formula to obtain a first irradiance data set:

[0163]

[0164] wherein, S represents the second irradiance data in the second irradiance data set; S t represents the first irradiance data in the first irradiance data set; S max represents the maximum irradiance data in the first irradiance data set; S min represents the minimum irradiance data in the first irradiance data set.

[0165] In one possible implementation, the clustering unit 302 is specifically configured to:

[0166] The elbow method is used to determine the number of clustering centers corresponding to the photovoltaic power data in the photovoltaic power data set, to obtain a first number.

[0167] The K-means clustering method is used according to the first number to cluster the photovoltaic power data in the photovoltaic power data set into K clusters, to obtain K second photovoltaic power data sets.

[0168] In one possible implementation, the first removing unit 303 is specifically configured to:

[0169] The photovoltaic power data matrix is constructed according to the target first photovoltaic power data set, to obtain a first photovoltaic power matrix, wherein the target first photovoltaic power data set is any one of the K first photovoltaic power data sets.

[0170] The photovoltaic power average value matrix is calculated according to the first photovoltaic power matrix, to obtain a photovoltaic power average value matrix.

[0171] obtaining a target photovoltaic power average value according to the photovoltaic power average value matrix;

[0172] obtaining a reference first photovoltaic data set by removing target first photovoltaic data in the target first photovoltaic data set, whose absolute value corresponding to a difference between the target first photovoltaic data and the target photovoltaic power average value is greater than an allowable threshold value;

[0173] repeating the construction of the photovoltaic power data matrix according to the target first photovoltaic power data set until the reference first photovoltaic data set is obtained by removing the target first photovoltaic data in the target first photovoltaic data set, whose absolute value corresponding to a difference between the target first photovoltaic data and the target photovoltaic power average value is greater than the allowable threshold value, until K reference first photovoltaic data sets corresponding to K first photovoltaic power data sets are obtained;

[0174] determining a union of the K reference first photovoltaic data sets as a second photovoltaic power data set.

[0175] In one possible implementation, the first calculation unit 304 is specifically configured to:

[0176] constructing a photovoltaic power and illumination intensity time series matrix corresponding to the second photovoltaic power data set;

[0177] calculating, according to the photovoltaic power and illumination intensity time series matrix, a Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set to obtain a Euclidean distance set;

[0178] determining a dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set to obtain a dynamic time warping distance set;

[0179] constructing a similarity function according to the Euclidean distance set and the dynamic time warping distance set;

[0180] calculating, according to the Euclidean distance set and the dynamic time warping distance set, a similarity between each second photovoltaic power data in the second photovoltaic power data set by the similarity function to obtain a similarity set.

[0181] In one possible implementation, the first removing unit 303 is specifically configured to:

[0182] constructing a photovoltaic power matrix according to a target first photovoltaic power data set, the target first photovoltaic power data set being any one of K first photovoltaic power data sets;

[0183] According to the first photovoltaic power matrix, an average value of photovoltaic power corresponding to each time period is calculated to obtain a photovoltaic power average value matrix;

[0184] According to the photovoltaic power average value matrix, a target photovoltaic power average value is obtained.

[0185] The target first photovoltaic data set is pruned to obtain a reference first photovoltaic data set, wherein the target first photovoltaic data corresponding to an absolute value greater than an allowable threshold is removed from the target first photovoltaic data set.

[0186] The above steps of constructing the photovoltaic power data matrix according to the target first photovoltaic power data set are repeatedly executed until the target first photovoltaic data corresponding to an absolute value greater than an allowable threshold is removed from the target first photovoltaic data set to obtain a reference first photovoltaic data set, until K first photovoltaic power data sets correspond to K reference first photovoltaic data sets, respectively.

[0187] The union of the K reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets is determined as a second photovoltaic power data set.

[0188] In one possible implementation, the first calculation unit 304 is specifically configured to:

[0189] A photovoltaic power and light intensity time sequence matrix corresponding to the second photovoltaic power data set is constructed.

[0190] According to the photovoltaic power and light intensity time sequence matrix, an Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain an Euclidean distance set.

[0191] A dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set is determined to obtain a dynamic time warping distance set.

[0192] A similarity function is constructed according to the Euclidean distance set and the dynamic time warping distance set.

[0193] According to the Euclidean distance set and the dynamic time warping distance set, a similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated by the similarity function to obtain a similarity set.

[0194] The embodiments of the present application also provide a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program causes a computer to execute part or all steps of any one of the distributed photovoltaic power anomaly identification methods described in the above method embodiments.

[0195] The embodiment of the present application further provides a computer program product, which comprises a non-transitory computer-readable storage medium storing a computer program, and the computer program causes a computer to execute some or all of the steps of any of the distributed photovoltaic power anomaly identification methods described in the above method embodiments.

[0196] It should be noted that, for the above-mentioned method embodiments, in order to simply describe, they are all described as a combination of a series of actions, but those skilled in the art should know that the present application is not limited to the action sequence described, because according to the present application, some steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0197] In the above embodiments, the description of each embodiment is focused on, and the part not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0198] In several embodiments provided by the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the units is only a logical function division. There can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection can be indirect coupling or communication connection through some interface, device or unit, and can be electrical or other form.

[0199] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0200] In addition, each functional unit in each embodiment of the application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software program module.

[0201] If the integrated unit is realized in the form of a software program module and sold or used as an independent product, it can be stored in a computer readable memory. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a memory and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned memory includes: a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0202] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by a program instructing relevant hardware, and the program can be stored in a computer readable memory, which can include a flash disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.

[0203] The embodiments of the present application are described in detail above, and the principles and implementation manners of the present application are described by applying specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea; meanwhile, for those of ordinary skill in the art, according to the idea of the present application, the specific implementation manner and application range will be changed, and the above description of the embodiments should not be understood as a limitation of the present application.

Claims

1. A distributed photovoltaic power anomaly discrimination method, characterized in that, The method comprises: The collected photovoltaic power data set is normalized to obtain a first light intensity data set; According to the first light intensity data set, the photovoltaic power data set is clustered to obtain K first photovoltaic power data sets; According to the K first photovoltaic power data sets, the abnormal data in the corresponding first photovoltaic power data set is removed to obtain a second photovoltaic power data set; The similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set; According to the similarity set, the profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a profile coefficient set; According to the profile coefficient set, the abnormal data in the second photovoltaic power data set is removed to obtain a fourth photovoltaic power data set.

2. The distributed photovoltaic power anomaly discrimination method according to claim 1, characterized in that, The collected photovoltaic power data set is normalized to obtain a first light intensity data set, comprising: The collected photovoltaic power data set is normalized to obtain a first light intensity data set, comprising: wherein represents second illumination intensity data in the second set of illumination intensity data; S t represents first illumination intensity data in the first set of illumination intensity data; S max represents maximum illumination intensity data in the first set of illumination intensity data; S min represents minimum illumination intensity data in the first set of illumination intensity data.

3. The method of claim 2, wherein, The elbow method is used to determine the number of clustering centers corresponding to the photovoltaic power data in the photovoltaic power data set to obtain a first number; According to the first number, the K-means clustering method is used to cluster the photovoltaic power data in the photovoltaic power data set into K clusters to obtain K second photovoltaic power data sets. According to the K first photovoltaic power data sets, the abnormal data in the K first photovoltaic power data sets is removed to obtain a second photovoltaic power data set, comprising:

4. The distributed photovoltaic power anomaly discrimination method according to claim 3, characterized in that, According to the target first photovoltaic power data set, a photovoltaic power data matrix is constructed to obtain a first photovoltaic power matrix, wherein the target first photovoltaic power data set is any one of the K first photovoltaic power data sets; According to the first photovoltaic power matrix, the photovoltaic power average value corresponding to each time period is calculated to obtain a photovoltaic power average value matrix; According to the photovoltaic power average value matrix, a target photovoltaic power average value is obtained; The target first photovoltaic data set is removed, and the absolute value corresponding to the difference between the target photovoltaic power average value is greater than the allowed threshold value to obtain a reference first photovoltaic data set; The above steps of constructing a photovoltaic power data matrix according to the target first photovoltaic power data set to obtain a first photovoltaic power matrix are repeatedly executed until the target first photovoltaic data set is removed, and the absolute value corresponding to the difference between the target photovoltaic power average value is greater than the allowed threshold value to obtain a reference first photovoltaic data set, until K first photovoltaic power data sets are obtained. ​ A union of reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets is determined as a second photovoltaic power data set.

5. The method of claim 1-4, wherein, The similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated to obtain a similarity set, including: A photovoltaic power and illumination intensity time series matrix corresponding to the second photovoltaic power data set is constructed; The Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set is calculated according to the photovoltaic power and illumination intensity time series matrix to obtain a Euclidean distance set; A dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set is determined to obtain a dynamic time warping distance set; A similarity function is constructed according to the Euclidean distance set and the dynamic time warping distance set; The similarity between each second photovoltaic power data in the second photovoltaic power data set is calculated by the similarity function according to the Euclidean distance set and the dynamic time warping distance set to obtain a similarity set.

6. A distributed photovoltaic power anomaly discrimination apparatus characterized by comprising: The device comprises: A processing unit configured to normalize illumination intensity data corresponding to photovoltaic power data in a collected photovoltaic power data set to obtain a first illumination intensity data set; A clustering unit configured to cluster the photovoltaic power data in the photovoltaic power data set according to the first illumination intensity data set to obtain K first photovoltaic power data sets; A first elimination unit configured to eliminate abnormal data in a corresponding first photovoltaic power data set according to an allowable threshold corresponding to the K first photovoltaic power data sets to obtain a second photovoltaic power data set; A first calculation unit configured to calculate the similarity between each second photovoltaic power data in the second photovoltaic power data set to obtain a similarity set; A second calculation unit configured to calculate a profile coefficient corresponding to each second photovoltaic power data in the second photovoltaic power data set according to the similarity set to obtain a profile coefficient set; A second elimination unit configured to eliminate abnormal data in the second photovoltaic power data set according to the profile coefficient set to obtain a fourth photovoltaic power data set.

7. The distributed photovoltaic power anomaly discrimination apparatus according to claim 6, characterized by, The first elimination unit is specifically configured to: construct a photovoltaic power matrix according to a target first photovoltaic power data set to obtain a first photovoltaic power matrix, the target first photovoltaic power data set being any one of the K first photovoltaic power data sets; calculate a photovoltaic power average value corresponding to each time period according to the first photovoltaic power matrix to obtain a photovoltaic power average value matrix; obtain a target photovoltaic power average value according to the photovoltaic power average value matrix; eliminate target first photovoltaic data with an absolute value greater than an allowable threshold between the target first photovoltaic data and the target photovoltaic power average value to obtain a reference first photovoltaic data set; The above-mentioned construction of the photovoltaic power data matrix according to the target first photovoltaic power data set is repeatedly performed to obtain a first photovoltaic power matrix, until the step of removing the target first photovoltaic data in the target first photovoltaic data set corresponding to the absolute value greater than the allowed threshold value between the target photovoltaic power average value is obtained to obtain the reference first photovoltaic data set, until K first photovoltaic power data sets correspond to the reference first photovoltaic data set respectively; The union of the reference first photovoltaic data sets corresponding to the K first photovoltaic power data sets is determined as a second photovoltaic power data set.

8. The distributed photovoltaic power anomaly discrimination apparatus according to claim 6, characterized by, The first calculation unit is specifically used for: constructing a photovoltaic power and illumination intensity time sequence matrix corresponding to the second photovoltaic power data set; calculating the Euclidean distance between each second photovoltaic power data in the second photovoltaic power data set according to the photovoltaic power and illumination intensity time sequence matrix to obtain a Euclidean distance set; determining a dynamic time warping distance corresponding to each Euclidean distance in the Euclidean distance set to obtain a dynamic time warping distance set; constructing a similarity function according to the Euclidean distance set and the dynamic time warping distance set; calculating the similarity between each second photovoltaic power data in the second photovoltaic power data set through the similarity function according to the Euclidean distance set and the dynamic time warping distance set to obtain a similarity set.

9. A terminal, characterized by comprising: The computer readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by the processor, the processor executes the distributed photovoltaic power anomaly identification method according to any one of claims 1-5.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by the processor, the processor executes the distributed photovoltaic power anomaly identification method according to any one of claims 1-5.

Images