Household photovoltaic power generation data anomaly detection method and system

Abstract

The invention relates to the technical field of household photovoltaic anomaly detection, and particularly discloses a household photovoltaic power generation data anomaly detection method and system, and the method comprises the steps: obtaining the environment coefficient of the surrounding of a household photovoltaic, and carrying out the environment partitioning of each household photovoltaic according to the environment coefficient of the surrounding of the household photovoltaic; obtaining the generating capacity of the household photovoltaic in each environment partition, determining a power generation fluctuation coefficient according to the generating capacity of the household photovoltaic in each environment partition, and screening out the benchmark household photovoltaic according to the power generation fluctuation coefficient; according to the benchmark household photovoltaic, the deviation degree between the rest household photovoltaic in the environment partition and the benchmark household photovoltaic is determined, and according to the deviation degree between the rest household photovoltaic in the environment partition and the benchmark household photovoltaic, the low-efficiency household photovoltaic is screened out; and obtaining low-efficiency household photovoltaic operation parameters, and determining a household photovoltaic operation and maintenance strategy according to the low-efficiency household photovoltaic operation parameters. According to the invention, environment partition is carried out on the environment around the household photovoltaic, the benchmarking household photovoltaic can be accurately found out, and the detection of the abnormally low-efficiency household photovoltaic is realized.

Description

Technical Field

[0001] The present application relates to the technical field of household photovoltaic anomaly detection, and more specifically, to a household photovoltaic power generation data anomaly detection method and system. Background Art

[0002] , photovoltaic power generation has been widely promoted and applied for its pollution-free and sustainable characteristics. Photovoltaic power generation in China is currently divided into two forms: centralized and distributed. The centralized form mainly appears in the form of photovoltaic power stations, while the distributed form mainly appears in the form of household photovoltaics. Among them, photovoltaic power stations are generally equipped with centralized monitoring systems due to their large capacity and highly concentrated geographical location, which can complete information interaction, intelligent operation and maintenance, and dispatching management in an integrated manner. Household photovoltaics are installed by residents themselves, and photovoltaic modules are installed in places with sufficient sunshine in their own homes (usually rooftops), and are equipped with inverters to invert and boost the DC output of photovoltaic modules to synchronize with the power grid. At the same time, smart meters are installed on the output side of the inverter to record the electricity generated by photovoltaics. Residential users can use this electricity for daily household electricity consumption, or they can sell it to power grid companies to make profits.

[0003] After photovoltaics have been widely deployed across the country and have become increasingly stable, how to accurately divide the environmental zones for household photovoltaics and promptly detect inefficient household photovoltaics with abnormal power generation, and improve the power generation efficiency of household photovoltaics, has become the key to the healthy development of photovoltaics. Summary of the invention

[0004] The present invention provides a household photovoltaic power generation data anomaly detection method and system, which is used to solve the problem that inefficient household photovoltaic power generation anomalies cannot be detected in time in the prior art, resulting in low power generation efficiency of household photovoltaic power generation, including:

[0005] Obtain the environmental coefficients around household photovoltaics, and divide the environmental zones of each household photovoltaic system according to the environmental coefficients around household photovoltaics;

[0006] Obtain the power generation of household photovoltaics in each environmental zone, determine the power generation fluctuation coefficient based on the power generation of household photovoltaics in each environmental zone, and screen out benchmark household photovoltaics based on the power generation fluctuation coefficient;

[0007] Determine the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics according to the benchmark household photovoltaics, and screen out inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics;

[0008] Obtain the operating parameters of inefficient household photovoltaics, and determine the household photovoltaic operation and maintenance strategy based on the operating parameters of inefficient household photovoltaics.

[0009] Furthermore, the environmental zoning of each household photovoltaic is carried out according to the environmental coefficient around the household photovoltaic, including:

[0010] Obtain the ambient light intensity around the household photovoltaic system, set the initial k value, and establish an ambient light intensity data set based on the ambient light intensity around the household photovoltaic system;

[0011] Determine the initial cluster center according to the initial k value, and calculate the Euclidean distance from the light intensity data in the ambient light intensity data set to the initial cluster center;

[0012] Divide each light intensity data into corresponding partitions according to the Euclidean distance from the light intensity data to the initial cluster center;

[0013] Update the cluster center according to the average value of the light intensity data in each partition, and iterate the new cluster center repeatedly until the cluster center no longer changes;

[0014] Calculating the DBI index of the clustering result, and determining whether the DBI index is less than a first preset threshold;

[0015] If the DBI index is greater than the first preset threshold, the initial k value is increased by one, and the ambient light intensity data set is clustered again until the DBI index is less than or equal to the first preset threshold. The clustering result is output to obtain the environmental zoning of household photovoltaics.

[0016] Furthermore, the DBI index of the clustering results is calculated, including:

[0017] The DBI index of the clustering results is calculated according to the DBI index calculation formula. The specific DBI index calculation formula is:

[0018]

[0019] Among them, D DBI is the DBI index, s i ,s j is the average distance from all data in the i, j partition to the cluster center, d i,j is the distance from the i-th cluster center to the j-th cluster center.

[0020] Furthermore, the power generation fluctuation coefficient is determined according to the power generation of household photovoltaics in each environmental zone, and benchmark household photovoltaics are screened out according to the power generation fluctuation coefficient, including:

[0021] Obtaining a power generation curve of household photovoltaics in each environmental zone, and screening out power generation valley points in the power generation curve that are less than a second preset threshold;

[0022] Obtain the descending curve on the left side of the valley point of the power generation curve, and calculate the standard deviation of the descending curve;

[0023] Determine the power generation fluctuation coefficient according to the standard deviation of the descending curve, obtain the photovoltaic panel temperature value of the household photovoltaic during the period corresponding to the descending curve, and correct the power generation fluctuation coefficient according to the photovoltaic panel temperature value;

[0024] According to the corrected power generation fluctuation coefficient, the household photovoltaic system with the smallest power generation fluctuation coefficient in each environmental zone is screened out, and the household photovoltaic system with the smallest power generation fluctuation coefficient is set as the benchmark household photovoltaic system.

[0025] Furthermore, the power generation fluctuation coefficient is corrected according to the temperature value of the photovoltaic panel, including:

[0026] According to the temperature value of the photovoltaic panel of the household photovoltaic in the corresponding period of the descending section curve, the difference between the temperature value of the photovoltaic panel in the descending section and the preset standard temperature value is determined. According to the difference between the temperature value of the photovoltaic panel in the descending section and the preset standard temperature value, the power generation fluctuation coefficient is corrected based on the power generation fluctuation coefficient correction formula. The power generation fluctuation coefficient correction formula is specifically as follows:

[0027]

[0028] Among them, E c is the corrected power generation fluctuation coefficient, E is the power generation fluctuation coefficient, T s is the preset standard temperature difference, T is the difference between the photovoltaic panel temperature value in the descending section and the preset standard temperature value, R is the preset range coefficient, and exp is the natural exponential function.

[0029] Furthermore, the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics is determined based on the benchmark household photovoltaics, including:

[0030] Obtain the power generation fluctuation coefficient of the benchmark household photovoltaic in the environmental zone, calculate the difference between the power generation fluctuation coefficients of the remaining household photovoltaics and the benchmark household photovoltaic, and obtain the power generation efficiency difference;

[0031] The geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone are obtained, and the power generation efficiency difference is corrected according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone. The degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone is determined according to the corrected power generation efficiency difference.

[0032] Furthermore, the power generation efficiency difference is corrected according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, including:

[0033] The coordinate distance between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone is calculated according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone;

[0034] Obtain the preset standard coordinate distance, calculate the difference between the coordinate distance of the remaining household photovoltaics and the benchmark household photovoltaics and the preset standard coordinate distance, and obtain the household photovoltaic distance difference;

[0035] The distance differences of all household photovoltaics in the environmental zone are normalized, and the power generation efficiency difference is corrected according to the normalized distance differences of all household photovoltaics.

[0036] Furthermore, the inefficient household photovoltaics are screened out according to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics, including:

[0037] According to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics, each household photovoltaic is ranked from high to low, and the household photovoltaics with a ranking greater than the third preset threshold are screened out according to the ranking results, and the household photovoltaics with a ranking greater than the third preset threshold are determined as inefficient household photovoltaics.

[0038] Furthermore, the operation and maintenance strategy of household photovoltaics is determined according to the operating parameters of the inefficient household photovoltaics, including:

[0039] Obtain the historical operating parameters of inefficient household photovoltaics and the corresponding household photovoltaic operation and maintenance strategies, and establish an operation and maintenance strategy data set based on the historical operating parameters of inefficient household photovoltaics and the corresponding household photovoltaic operation and maintenance strategies;

[0040] Establish an initial operation and maintenance strategy model according to the operation and maintenance strategy data set and train the initial operation and maintenance strategy model to obtain a trained operation and maintenance strategy model;

[0041] The operating parameters of the current inefficient household photovoltaics are input into the trained operation and maintenance strategy model, and the corresponding household photovoltaic operation and maintenance strategy is output.

[0042] In order to achieve the above object, the present invention also provides a household photovoltaic power generation data anomaly detection system, comprising:

[0043] The environmental zoning module is used to obtain the environmental coefficients around household photovoltaics and to divide the environmental zoning of each household photovoltaic according to the environmental coefficients around the household photovoltaics;

[0044] The benchmark screening module is used to obtain the power generation of household photovoltaics in each environmental zone, determine the power generation fluctuation coefficient according to the power generation of household photovoltaics in each environmental zone, and screen out benchmark household photovoltaics according to the power generation fluctuation coefficient;

[0045] The inefficient screening module is used to determine the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics according to the benchmark household photovoltaics, and screen out the inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics;

[0046] The strategy setting module is used to obtain the operating parameters of the inefficient household photovoltaic system and determine the household photovoltaic operation and maintenance strategy according to the operating parameters of the inefficient household photovoltaic system.

[0047] The beneficial effects of the present invention are:

[0048] By applying the above technical scheme, the present invention performs environmental zoning for each household photovoltaic system according to the environmental coefficient around the household photovoltaic system, and the zoning result is more accurate, scientific and reasonable, so as to quickly find the benchmark household photovoltaic system in each environmental zone, and screen out the inefficient household photovoltaic system with abnormal power generation according to the benchmark household photovoltaic system, so as to timely discover the abnormality of household photovoltaic system, facilitate the subsequent maintenance and inspection of the abnormal household photovoltaic system, and greatly improve the power generation efficiency of household photovoltaic system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 The overall flow chart of a method for detecting abnormality of household photovoltaic power generation data proposed in an embodiment of the present invention is shown;

[0051] Figure 2 A schematic structural diagram of a household photovoltaic power generation data anomaly detection system proposed in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] The present application embodiment provides a method for detecting abnormality in household photovoltaic power generation data, such as Figure 1 As shown, including:

[0054] S101, obtaining the environmental coefficient of the household photovoltaic system, and performing environmental zoning for each household photovoltaic system according to the environmental coefficient of the household photovoltaic system;

[0055] In some embodiments of the present application, each household photovoltaic is environmentally zoned according to the environmental coefficient around the household photovoltaic, including: obtaining the ambient light intensity around the household photovoltaic, setting an initial k value, and establishing an ambient light intensity data set according to the ambient light intensity around the household photovoltaic; determining an initial clustering center according to the initial k value, and calculating the Euclidean distance from the light intensity data in the ambient light intensity data set to the initial clustering center; dividing each light intensity data into corresponding partitions according to the Euclidean distance from the light intensity data to the initial clustering center; updating the clustering center according to the average value of the light intensity data in each partition, and repeatedly iterating the new clustering center until the clustering center no longer changes; calculating the DBI index of the clustering result, and judging whether the DBI index is less than a first preset threshold; if the DBI index is greater than the first preset threshold, adding one to the initial k value, re-clustering the ambient light intensity data set until the DBI index is less than or equal to the first preset threshold, outputting the clustering result, and obtaining the environmental zoning of the household photovoltaic.

[0056] In some embodiments of the present application, calculating the DBI index of the clustering result includes:

[0057] The DBI index of the clustering results is calculated according to the DBI index calculation formula. The specific DBI index calculation formula is:

[0058]

[0059] Among them, D DBI is the DBI index, s i ,s j is the average distance from all data in the i, j partition to the cluster center, d i,j is the distance from the i-th cluster center to the j-th cluster center.

[0060] In this embodiment, the initial k value is set to 2, the maximum k value is set to 10, and the ambient light intensity data of household photovoltaics are clustered based on the k-means clustering algorithm. The optimal k value is calculated according to the DBI index, so that the distance between the final environmental partitions is larger and the distance within the partition is smaller, so as to achieve a better clustering effect.

[0061] S102, obtaining the power generation of household photovoltaics in each environmental zone, determining the power generation fluctuation coefficient according to the power generation of household photovoltaics in each environmental zone, and selecting benchmark household photovoltaics according to the power generation fluctuation coefficient;

[0062] In some embodiments of the present application, a power generation fluctuation coefficient is determined based on the power generation of household photovoltaics in each environmental zone, and a benchmark household photovoltaic is screened out based on the power generation fluctuation coefficient, including: obtaining the power generation curve of household photovoltaics in each environmental zone, and screening out the power generation valley points in the power generation curve that are less than a second preset threshold; obtaining the descending section curve on the left side of the power generation valley point in the power generation curve, and calculating the standard deviation of the descending section curve; determining the power generation fluctuation coefficient based on the standard deviation of the descending section curve, obtaining the photovoltaic panel temperature value of the household photovoltaics in the corresponding period of the descending section curve, and correcting the power generation fluctuation coefficient based on the photovoltaic panel temperature value; screening out the household photovoltaics with the smallest power generation fluctuation coefficient in each environmental zone based on the corrected power generation fluctuation coefficient, and setting the household photovoltaics with the smallest power generation fluctuation coefficient as the benchmark household photovoltaics.

[0063] In this embodiment, a descending curve in which the power generation of household photovoltaic power generation drops from a peak point to a valley point less than a second preset threshold value is screened out, and the standard deviation of the descending curve is set as the power generation fluctuation coefficient, thereby screening out a benchmark household photovoltaic with the smallest power generation fluctuation coefficient.

[0064] In some embodiments of the present application, the power generation fluctuation coefficient is corrected according to the photovoltaic panel temperature value, including: determining the difference between the photovoltaic panel temperature value of the descending section and the preset standard temperature value according to the photovoltaic panel temperature value of the household photovoltaic panel in the period corresponding to the descending section curve, and correcting the power generation fluctuation coefficient based on the power generation fluctuation coefficient correction formula according to the difference between the photovoltaic panel temperature value of the descending section and the preset standard temperature value. The power generation fluctuation coefficient correction formula is specifically as follows:

[0065]

[0066] Among them, E c is the corrected power generation fluctuation coefficient, E is the power generation fluctuation coefficient, T s is the preset standard temperature difference, T is the difference between the photovoltaic panel temperature value in the descending section and the preset standard temperature value, R is the preset range coefficient, and exp is the natural exponential function.

[0067] In this embodiment, the power generation fluctuation coefficient is corrected by the difference between the temperature value of the photovoltaic panel in the descending section and the preset standard temperature value, thereby reducing the influence of the ambient temperature on the power generation fluctuation coefficient, thereby making the power generation fluctuation coefficient more accurate.

[0068] S103, determining the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics according to the benchmark household photovoltaics, and screening out inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics;

[0069] In some embodiments of the present application, the degree of deviation of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone is determined based on the benchmark household photovoltaics, including: obtaining the power generation fluctuation coefficient of the benchmark household photovoltaics in the environmental zone, calculating the difference between the power generation fluctuation coefficients of the remaining household photovoltaics and the benchmark household photovoltaics, and obtaining the power generation efficiency difference; obtaining the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, correcting the power generation efficiency difference according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, and determining the degree of deviation of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone according to the corrected power generation efficiency difference.

[0070] In some embodiments of the present application, the power generation efficiency difference is corrected according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, including: calculating the coordinate distance between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone; obtaining the preset standard coordinate distance, calculating the difference between the coordinate distance between the remaining household photovoltaics and the benchmark household photovoltaics and the preset standard coordinate distance, and obtaining the household photovoltaic distance difference; normalizing all household photovoltaic distance differences in the environmental zone, and correcting the power generation efficiency difference according to the normalized household photovoltaic distance difference.

[0071] In this embodiment, the household photovoltaic distance difference is normalized to limit the range of the household photovoltaic distance difference to [0, 1], and the normalized household photovoltaic distance difference is multiplied by the power generation efficiency difference to correct the power generation efficiency difference. The corrected power generation efficiency difference is used as the deviation degree between the household photovoltaic and the benchmark household photovoltaic.

[0072] S104, obtaining operating parameters of the inefficient household photovoltaic system, and determining a household photovoltaic operation and maintenance strategy according to the operating parameters of the inefficient household photovoltaic system.

[0073] In some embodiments of the present application, inefficient household photovoltaics are screened out according to the degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, including: sorting each household photovoltaic in a descending order according to the degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, and screening out household photovoltaics whose ranking is greater than a third preset threshold according to the sorting results, and determining the household photovoltaics whose ranking is greater than the third preset threshold as inefficient household photovoltaics.

[0074] In this embodiment, household photovoltaics with a low deviation ranking are screened out to obtain inefficient household photovoltaics, which facilitates the subsequent setting of operation and maintenance strategies for the inefficient household photovoltaics.

[0075] In some embodiments of the present application, a household photovoltaic operation and maintenance strategy is determined based on the operating parameters of inefficient household photovoltaics, including: obtaining historical operating parameters of inefficient household photovoltaics and corresponding household photovoltaic operation and maintenance strategies, and establishing an operation and maintenance strategy data set based on the historical operating parameters of inefficient household photovoltaics and corresponding household photovoltaic operation and maintenance strategies; establishing an initial operation and maintenance strategy model based on the operation and maintenance strategy data set and training the initial operation and maintenance strategy model to obtain a trained operation and maintenance strategy model; inputting the current operating parameters of inefficient household photovoltaics into the trained operation and maintenance strategy model, and outputting the corresponding household photovoltaic operation and maintenance strategy.

[0076] In this embodiment, the operating parameters of inefficient household photovoltaics include the rated power and temperature coefficient of the photovoltaic equipment, the average sunshine duration and sunshine intensity distribution around the photovoltaic equipment, etc. An operation and maintenance strategy model is established based on a deep learning neural network, and the operation and maintenance strategy model is trained according to historical operating parameters and corresponding historical operation and maintenance strategies of household photovoltaics. The trained operation and maintenance strategy model outputs corresponding operation and maintenance strategies for the current operating parameters of inefficient household photovoltaics, thereby effectively improving the power generation efficiency of household photovoltaics.

[0077] Based on the same technical concept, such as Figure 2 As shown, the present invention also provides a household photovoltaic power generation data anomaly detection system, comprising:

[0078] The environmental zoning module is used to obtain the environmental coefficients around household photovoltaics, and to divide the environmental zoning of each household photovoltaic according to the environmental coefficients around the household photovoltaics; the benchmark screening module is used to obtain the power generation of household photovoltaics in each environmental zone, determine the power generation fluctuation coefficient according to the power generation of household photovoltaics in each environmental zone, and screen out benchmark household photovoltaics according to the power generation fluctuation coefficient; the inefficiency screening module is used to determine the degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone according to the benchmark household photovoltaics, and screen out inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone; the strategy setting module is used to obtain the operating parameters of inefficient household photovoltaics, and determine the household photovoltaic operation and maintenance strategy according to the operating parameters of inefficient household photovoltaics.

[0079] By applying the above technical scheme, the present invention obtains the environmental coefficient around household photovoltaics, and performs environmental zoning for each household photovoltaic according to the environmental coefficient around household photovoltaics; obtains the power generation of household photovoltaics in each environmental zone, determines the power generation fluctuation coefficient according to the power generation of household photovoltaics in each environmental zone, and selects benchmark household photovoltaics according to the power generation fluctuation coefficient; determines the degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone according to the benchmark household photovoltaics, and selects inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone; obtains the operating parameters of inefficient household photovoltaics, and determines the operation and maintenance strategy of household photovoltaics according to the operating parameters of inefficient household photovoltaics. By performing environmental zoning for the environment around household photovoltaics, benchmark household photovoltaics can be accurately found, and abnormally inefficient household photovoltaics can be detected.

[0080] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting abnormality in household photovoltaic power generation data, characterized in that: include: Obtain the environmental coefficients around household photovoltaics, and divide the environmental zones of each household photovoltaic system according to the environmental coefficients around household photovoltaics; Obtain the power generation of household photovoltaics in each environmental zone, determine the power generation fluctuation coefficient based on the power generation of household photovoltaics in each environmental zone, and screen out benchmark household photovoltaics based on the power generation fluctuation coefficient; Determine the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics according to the benchmark household photovoltaics, and screen out inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics; Obtain the operating parameters of inefficient household photovoltaics, and determine the household photovoltaic operation and maintenance strategy based on the operating parameters of inefficient household photovoltaics.

2. The method for detecting abnormality of household photovoltaic power generation data according to claim 1, characterized in that: Environmental zoning of each household photovoltaic system is carried out according to the environmental coefficients around the household photovoltaic system, including: Obtain the ambient light intensity around the household photovoltaic system, set the initial k value, and establish an ambient light intensity data set based on the ambient light intensity around the household photovoltaic system; Determine the initial cluster center according to the initial k value, and calculate the Euclidean distance from the light intensity data in the ambient light intensity data set to the initial cluster center; Divide each light intensity data into corresponding partitions according to the Euclidean distance from the light intensity data to the initial cluster center; Update the cluster center according to the average value of the light intensity data in each partition, and iterate the new cluster center repeatedly until the cluster center no longer changes; Calculating the DBI index of the clustering result, and determining whether the DBI index is less than a first preset threshold; If the DBI index is greater than the first preset threshold, the initial k value is increased by one, and the ambient light intensity data set is clustered again until the DBI index is less than or equal to the first preset threshold. The clustering result is output to obtain the environmental zoning of household photovoltaics.

3. The method for detecting abnormality of household photovoltaic power generation data according to claim 2, characterized in that: Calculate the DBI index of the clustering results, including: The DBI index of the clustering results is calculated according to the DBI index calculation formula. The specific DBI index calculation formula is: Among them, D DBI is the DBI index, s i ,s j is the average distance from all data in the i, j partition to the cluster center, d i,j is the distance from the i-th cluster center to the j-th cluster center.

4. The method for detecting abnormality of household photovoltaic power generation data according to claim 1, characterized in that: The power generation fluctuation coefficient is determined according to the power generation of household photovoltaics in each environmental zone, and the benchmark household photovoltaics are screened out according to the power generation fluctuation coefficient, including: Obtaining a power generation curve of household photovoltaics in each environmental zone, and screening out power generation valley points in the power generation curve that are less than a second preset threshold; Obtain the descending curve on the left side of the valley point of the power generation curve, and calculate the standard deviation of the descending curve; Determine the power generation fluctuation coefficient according to the standard deviation of the descending curve, obtain the photovoltaic panel temperature value of the household photovoltaic during the period corresponding to the descending curve, and correct the power generation fluctuation coefficient according to the photovoltaic panel temperature value; According to the corrected power generation fluctuation coefficient, the household photovoltaic system with the smallest power generation fluctuation coefficient in each environmental zone is screened out, and the household photovoltaic system with the smallest power generation fluctuation coefficient is set as the benchmark household photovoltaic system.

5. The method for detecting abnormality of household photovoltaic power generation data according to claim 4, characterized in that: Correct the power generation fluctuation coefficient according to the temperature value of the photovoltaic panel, including: According to the temperature value of the photovoltaic panel of the household photovoltaic in the corresponding period of the descending section curve, the difference between the temperature value of the photovoltaic panel in the descending section and the preset standard temperature value is determined. According to the difference between the temperature value of the photovoltaic panel in the descending section and the preset standard temperature value, the power generation fluctuation coefficient is corrected based on the power generation fluctuation coefficient correction formula. The power generation fluctuation coefficient correction formula is specifically as follows: Among them, E c is the corrected power generation fluctuation coefficient, E is the power generation fluctuation coefficient, T s is the preset standard temperature difference, T is the difference between the photovoltaic panel temperature value in the descending section and the preset standard temperature value, R is the preset range coefficient, and exp is the natural exponential function.

6. The method for detecting abnormality of household photovoltaic power generation data according to claim 5, characterized in that: According to the benchmark household photovoltaic system, the deviation degree between the remaining household photovoltaic systems in the environmental zone and the benchmark household photovoltaic system is determined, including: Obtain the power generation fluctuation coefficient of the benchmark household photovoltaic in the environmental zone, calculate the difference between the power generation fluctuation coefficients of the remaining household photovoltaics and the benchmark household photovoltaic, and obtain the power generation efficiency difference; The geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone are obtained, and the power generation efficiency difference is corrected according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone. The degree of deviation between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone is determined according to the corrected power generation efficiency difference.

7. The method for detecting abnormality of household photovoltaic power generation data according to claim 6, characterized in that: Correct the power generation efficiency difference based on the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone, including: The coordinate distance between the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone is calculated according to the geographical coordinates of the remaining household photovoltaics and the benchmark household photovoltaics in the environmental zone; Obtain the preset standard coordinate distance, calculate the difference between the coordinate distance of the remaining household photovoltaics and the benchmark household photovoltaics and the preset standard coordinate distance, and obtain the household photovoltaic distance difference; The distance differences of all household photovoltaics in the environmental zone are normalized, and the power generation efficiency difference is corrected according to the normalized distance differences of all household photovoltaics.

8. The method for detecting abnormality of household photovoltaic power generation data according to claim 1, characterized in that: Inefficient household photovoltaics are screened out based on the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics, including: According to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics, each household photovoltaic is ranked from high to low, and the household photovoltaics with a ranking greater than the third preset threshold are screened out according to the ranking results, and the household photovoltaics with a ranking greater than the third preset threshold are determined as inefficient household photovoltaics.

9. The method for detecting abnormality of household photovoltaic power generation data according to claim 8, characterized in that: Determine the operation and maintenance strategy of household photovoltaics based on the operating parameters of inefficient household photovoltaics, including: Obtain the historical operating parameters of inefficient household photovoltaics and the corresponding household photovoltaic operation and maintenance strategies, and establish an operation and maintenance strategy data set based on the historical operating parameters of inefficient household photovoltaics and the corresponding household photovoltaic operation and maintenance strategies; Establish an initial operation and maintenance strategy model according to the operation and maintenance strategy data set and train the initial operation and maintenance strategy model to obtain a trained operation and maintenance strategy model; The operating parameters of the current inefficient household photovoltaics are input into the trained operation and maintenance strategy model, and the corresponding household photovoltaic operation and maintenance strategy is output.

10. A household photovoltaic power generation data anomaly detection system, characterized in that: include: The environmental zoning module is used to obtain the environmental coefficients around household photovoltaics and to divide the environmental zoning of each household photovoltaic according to the environmental coefficients around the household photovoltaics; The benchmark screening module is used to obtain the power generation of household photovoltaics in each environmental zone, determine the power generation fluctuation coefficient according to the power generation of household photovoltaics in each environmental zone, and screen out benchmark household photovoltaics according to the power generation fluctuation coefficient; The inefficient screening module is used to determine the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics according to the benchmark household photovoltaics, and screen out the inefficient household photovoltaics according to the degree of deviation between the remaining household photovoltaics in the environmental zone and the benchmark household photovoltaics; The strategy setting module is used to obtain the operating parameters of the inefficient household photovoltaic system and determine the household photovoltaic operation and maintenance strategy according to the operating parameters of the inefficient household photovoltaic system.

Images