Solar cell panel fault analysis method and system

Abstract

The invention provides a solar cell panel fault analysis method and system, and belongs to the technical field of fault analysis, and the method comprises the steps: obtaining the operation data of a plurality of solar cell panels under all external conditions, and determining a combined external category and first operation data; analyzing the first operation data, and determining a fault type of each sub-combination external category in the combination external category and a fault feature vector of each fault sub-type in the fault types; real-time operation data of the solar cell panel are collected in real time, and a real-time feature vector is determined; determining the equipment state of the solar cell panel; and carrying out fault processing or maintenance on the solar cell panel based on the equipment state of the solar cell panel. The fault type of each sub-combination under the external type can be accurately identified, early identification and classification of faults are realized, fault processing and maintenance strategies are optimized, the operation efficiency and reliability of equipment are remarkably improved, the maintenance cost is reduced, and the service life of a solar cell panel is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of fault analysis, and in particular to a solar panel fault analysis method and system. Background Art

[0002] Traditional solar panel fault detection relies on static data and empirical judgment, which makes it difficult to achieve real-time monitoring and accurate fault identification. In addition, existing technologies often ignore the impact of external conditions on panel performance, resulting in insufficient accuracy in fault analysis. With the development of solar power generation technology, an innovative method is urgently needed to achieve comprehensive monitoring and rapid response to the status of solar panels.

[0003] Therefore, the present invention provides a solar panel fault analysis method and system. Summary of the invention

[0004] The present invention provides a solar panel fault analysis method and system, which determines first operation data through acquired operation data of the solar panel, analyzes the first operation data to determine the fault type of the combined external category and the fault feature vector, combines the real-time feature vector of the solar panel to determine the device state of the solar panel, and processes or maintains according to the device state, which can improve data processing efficiency, accurately identify the fault type under each sub-combination external category, realize early identification and classification of faults, optimize fault processing and maintenance strategies, significantly improve the operation efficiency and reliability of the equipment, reduce maintenance costs, and extend the service life of the solar panel.

[0005] In one aspect, the present invention provides a solar panel fault analysis method, comprising:

[0006] 101: Acquire operating data of a plurality of solar panels under all external conditions, pre-process the operating data to determine a combined external category, and determine first operating data;

[0007] 102: Analyze the first operation data to determine the fault type of each sub-combined external category in the combined external category and the fault feature vector of each fault sub-type in the fault type;

[0008] 103: collecting real-time operation data of the solar panel in real time, and analyzing the real-time operation data to determine a real-time feature vector;

[0009] 104: Determine a device state of the solar panel based on a real-time feature vector of the solar panel and fault features of all fault subtypes in the fault types of all sub-combined external categories;

[0010] 105: Based on the device status of the solar panel, perform troubleshooting or maintenance on the solar panel.

[0011] According to a solar panel fault analysis method provided by the present invention, operating data of multiple solar panels under all external conditions are obtained, the operating data are pre-processed to determine the combined external category, and the first operating data is determined, including:

[0012] Determine a grouping rule for each external condition, and determine an external classification of the corresponding external condition based on the grouping rule for each external condition;

[0013] Arrange and combine all external classifications to determine a combined external category, wherein the combined external category includes a plurality of sub-combined external categories;

[0014] Preprocessing the operation data includes data cleaning, data normalization and data grouping;

[0015] Allocate each piece of operation sub-data in the operation data after data cleaning and data normalization based on the combined external category, and determine the category sub-data and the number of category sub-data of each sub-combined external category class in the combined external category;

[0016] Determine that all sub-combination external categories whose category sub-data quantity of the sub-combination external category in the combined external category is 0 are false sub-combination external categories, eliminate all false sub-combination external categories in the combined external category, and perform a first update on the combined external category based on the elimination result;

[0017] Performing category merging on all sub-combined external categories in the combined external category after the first update, whose category sub-data quantity is less than a preset quantity threshold, and performing data merging on the category sub-data of the sub-combined external categories that are subjected to category merging, performing a second update on the combined external category based on the category merging result, and performing a third update on the category sub-data of each sub-combined external category in the combined external category based on the data merging result;

[0018] First operating data is determined based on the category sub-data of the combined ambient category after the first update, the second update, and all the sub-combined ambient categories after the third update.

[0019] According to a solar panel fault analysis method provided by the present invention, first operation data is analyzed to determine a fault type and a fault feature vector of each fault subtype in the fault type, including:

[0020] Based on the category sub-data of each sub-combination external category in the first operation data, determining all operation sub-data marked with faults in the category sub-data as sub-fault data of the corresponding sub-combination external category;

[0021] Extracting features of each piece of fault operation data in the sub-fault data of each sub-combination external category in the combined external category, and determining a sub-fault feature vector of each piece of fault operation data;

[0022] Performing cluster analysis on all fault operation data in the sub-fault data of each sub-combined external category in the combined external category;

[0023] Determine the fault type of each sub-combined external category in the combined external category based on all clusters of the cluster analysis result, wherein the fault type includes multiple fault sub-types;

[0024] Based on all fault operation data in each cluster group of each sub-combination external category, a fault feature vector of a fault sub-type corresponding to each cluster group in each sub-combination external category is determined.

[0025] According to a solar panel fault analysis method provided by the present invention, based on all fault operation data in each cluster group of each sub-combination external category, a fault feature vector of a fault sub-type corresponding to each cluster group in each sub-combination external category is determined, including:

[0026] Fa i =(f i1 …, f ij …, f iN1 );

[0027]

[0028] Among them, Fa i represents the fault feature vector of the fault subtype corresponding to the i-th cluster in the cluster analysis results, f i1 、f ij 、f iN1 They represent the feature values ​​of the 1st feature, the jth feature, and the N1th feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, respectively. N1 represents the number of features in the fault feature vector. ij represents the feature fitting value of the jth feature in the fault feature vector of the fault subtype corresponding to the i-th cluster family, Med(X ijk ) represents the feature median of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, Q3 ij 、Q1 ij They represent the third quartile and the first quartile of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, respectively, ij The characteristic skewness of the jth feature in the fault feature vector corresponding to the fault subtype of the i-th cluster family, δ ijThe characteristic kurtosis of the jth feature in the fault feature vector corresponding to the fault subtype of the i-th cluster family, σ ij represents the standard deviation of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, C ij represents the cluster center feature value of the jth feature of the corresponding fault subtype of the i-th cluster family, X ijk represents the eigenvalue of the jth feature in the sub-fault feature vector of the kth fault operation data in the fault subtype corresponding to the i-th cluster family, iN2 represents the number of fault operation data in the i-th cluster family, μ ij Represents the average value of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family.

[0029] According to a solar panel fault analysis method provided by the present invention, real-time operation data of the solar panel is collected in real time, and the real-time operation data is analyzed to determine the real-time feature vector, including:

[0030] Based on the sensor group, real-time operation data of the solar panel is collected in real time, wherein the sensor group includes at least a current sensor, a voltage sensor, a power sensor, a temperature sensor and a radiation intensity sensor, and the real-time operation data includes a plurality of real-time operation sub-data;

[0031] Based on the real-time operation sub-data corresponding to each sensor in the sensor group, feature extraction is performed on each real-time operation sub-data to determine a feature value of each real-time operation sub-data;

[0032] Extracting feature values ​​of all real-time operation sub-data based on external conditions, and determining sub-combination external categories of the real-time operation data based on all extracted feature values ​​and combined external categories;

[0033] A real-time feature vector is determined based on the feature values ​​of all real-time operation sub-data except the external conditions in the real-time operation data.

[0034] According to a solar panel fault analysis method provided by the present invention, based on the real-time feature vector of the solar panel and the fault features of all fault subtypes in the fault types of all sub-combined external categories, the device state of the solar panel is determined, including:

[0035] determining similarity of a real-time feature vector of the solar panel and each fault feature vector in a sub-combination external category of the real-time operation sub-data;

[0036] Determine a device state of the solar panel based on all similarity values, wherein the device state includes a faulty device, a device to be faulty, and a normal device;

[0037] If any similarity value is greater than a first fault preset threshold, the solar panel is determined to be a faulty device, and the real-time fault type is determined based on the fault subtypes corresponding to all similarity values ​​greater than the first fault preset threshold;

[0038] If any similarity value is between the first fault preset threshold and the second fault preset threshold, the solar panel is determined to be a device to be faulty, and the real-time fault type to be occurred is determined based on the fault subtypes corresponding to all similarity values ​​between the first fault preset threshold and the second fault preset threshold;

[0039] Otherwise, the solar panel is determined to be a normal device.

[0040] According to a solar panel fault analysis method provided by the present invention, fault processing or maintenance of the solar panel is performed based on the device status of the solar panel, including:

[0041] If the solar panel is a faulty device, a fault handling strategy is generated based on the real-time fault type and real-time operating characteristics of the solar panel;

[0042] If the solar panel is a device that is about to fail, a maintenance suggestion is generated based on the real-time type of failure that is about to fail of the solar panel;

[0043] Perform fault handling on solar panels based on fault handling strategies and perform equipment maintenance on solar panels based on maintenance recommendations;

[0044] A fault report is generated based on the real-time fault type, fault handling strategy and fault handling result of the solar panel, and a maintenance report is generated based on the real-time pending fault type, maintenance suggestion and maintenance result of the solar panel.

[0045] On the other hand, the present invention also provides a solar panel fault analysis system, comprising:

[0046] Acquisition module: acquires the operating data of multiple solar panels under all external conditions, pre-processes the operating data to determine the combined external categories, and first operates the data;

[0047] Analysis module: analyzing the first operation data, determining the fault type of each sub-combination external category in the combination external category and the fault feature vector of each fault sub-type in the fault type;

[0048] Acquisition module: collects real-time operation data of solar panels in real time, analyzes the operation data and determines the real-time feature vector;

[0049] Fault module: determines the device status of the solar panel based on the real-time feature vector of the solar panel and the fault features of all fault subtypes in the fault types of all sub-combined external categories;

[0050] Processing module: based on the device status of the solar panel, perform fault processing or maintenance on the solar panel.

[0051] Compared with the prior art, the beneficial effects of this application are as follows:

[0052] The first operating data is determined by obtaining the operating data of the solar panel, the fault type and the fault feature vector of the combined external category are determined by analyzing the first operating data, the device status of the solar panel is determined in combination with the real-time feature vector of the solar panel, and processing or maintenance is performed according to the device status. This can improve data processing efficiency, accurately identify the fault type under each sub-combination external category, achieve early identification and classification of faults, optimize fault processing and maintenance strategies, significantly improve the operating efficiency and reliability of the equipment, reduce maintenance costs, and extend the service life of the solar panel. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 The present invention is a schematic flow chart of a solar panel fault analysis method provided by an embodiment of the present invention.

[0055] Figure 2 It is a structural schematic diagram of a solar panel fault analysis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Embodiment 1:

[0058] The embodiment of the present invention provides a solar panel fault analysis method, such as Figure 1 As shown, including:

[0059] 101: Acquire operating data of a plurality of solar panels under all external conditions, pre-process the operating data to determine a combined external category, and determine first operating data;

[0060] 102: Analyze the first operation data to determine the fault type of each sub-combined external category in the combined external category and the fault feature vector of each fault sub-type in the fault type;

[0061] 103: collecting real-time operation data of the solar panel in real time, and analyzing the real-time operation data to determine a real-time feature vector;

[0062] 104: Determine a device state of the solar panel based on a real-time feature vector of the solar panel and fault features of all fault subtypes in the fault types of all sub-combined external categories;

[0063] 105: Based on the device status of the solar panel, perform troubleshooting or maintenance on the solar panel.

[0064] In this embodiment, the operating data of multiple solar panels under all external conditions are obtained, and the operating data of different solar panels under various environmental conditions (such as temperature, humidity, light intensity, etc.) are collected.

[0065] In this embodiment, the collected operating data is cleaned, normalized, and grouped to ensure the accuracy and consistency of the data.

[0066] In this embodiment, different external condition combination categories are defined based on the processed data to facilitate classification and analysis.

[0067] In this embodiment, the first operation data is deeply analyzed to identify the fault type of each sub-combination external category in the combined external category, and for each fault sub-type in the fault type, a corresponding fault feature vector is determined.

[0068] In this embodiment, the operating status of the solar panel is continuously monitored by the sensor group to collect current operating data.

[0069] In this embodiment, the real-time feature vector is compared with all fault feature vectors, and the device status of the solar panel is determined based on these data. The status can be normal, faulty, or about to occur.

[0070] In this embodiment, corresponding measures are taken according to the determined device status. If the solar panel is determined to be a faulty device, fault processing is performed; if it is a device that is about to fail, maintenance suggestions are implemented to prevent the failure from occurring.

[0071] The beneficial effects of the above technical solution are as follows: determining the first operating data through the acquired operating data of the solar panel, analyzing the first operating data to determine the fault type and fault feature vector of the combined external category, determining the device status of the solar panel in combination with the real-time feature vector of the solar panel, and performing processing or maintenance according to the device status, which can improve data processing efficiency, accurately identify the fault type under each sub-combination external category, realize early identification and classification of faults, optimize fault processing and maintenance strategies, significantly improve the operating efficiency and reliability of the equipment, reduce maintenance costs, and extend the service life of the solar panel.

[0072] Embodiment 2:

[0073] The embodiment of the present invention provides a solar panel fault analysis method, which obtains operating data of multiple solar panels under all external conditions, pre-processes the operating data to determine the combined external category, and determines the first operating data, including:

[0074] Determine a grouping rule for each external condition, and determine an external classification of the corresponding external condition based on the grouping rule for each external condition;

[0075] Arrange and combine all external classifications to determine a combined external category, wherein the combined external category includes a plurality of sub-combined external categories;

[0076] Preprocessing the operation data includes data cleaning, data normalization and data grouping;

[0077] Allocate each piece of operation sub-data in the operation data after data cleaning and data normalization based on the combined external category, and determine the category sub-data and the number of category sub-data of each sub-combined external category class in the combined external category;

[0078] Determine that all sub-combination external categories whose category sub-data quantity of the sub-combination external category in the combined external category is 0 are false sub-combination external categories, eliminate all false sub-combination external categories in the combined external category, and perform a first update on the combined external category based on the elimination result;

[0079] Performing category merging on all sub-combined external categories in the combined external category after the first update, whose category sub-data quantity is less than a preset quantity threshold, and performing data merging on the category sub-data of the sub-combined external categories that are subjected to category merging, performing a second update on the combined external category based on the category merging result, and performing a third update on the category sub-data of each sub-combined external category in the combined external category based on the data merging result;

[0080] First operating data is determined based on the category sub-data of the combined ambient category after the first update, the second update, and all the sub-combined ambient categories after the third update.

[0081] In this embodiment, all lighting conditions include all lighting intensities, all ambient temperatures, and all solar radiation intensities.

[0082] In this embodiment, the grouping rules for each external condition are determined, for example: time external condition: grouping by hour, day, week or other time intervals; light intensity external condition: divide the light intensity into several intervals, low light (0-300W / m 2 ), medium light (300-700W / m 2 ), high light (700W / m 2 Temperature range: The temperature range is defined according to the temperature distribution, low temperature (less than 10°C), moderate temperature (10°C-40°C), and high temperature (greater than 40°C).

[0083] In this embodiment, grouping criteria are formulated for each external condition (such as temperature, humidity, light intensity, etc.) so as to classify these conditions into specific external categories, and all external categories are combined to determine combined external categories. These combinations contain multiple sub-combination external categories that can cover different operating conditions.

[0084] In this embodiment, the preprocessing of row data includes the following steps: data cleaning removes erroneous or invalid data to ensure data quality; data normalization converts data of different dimensions into a unified standard; and data grouping classifies the cleaned and normalized data.

[0085] In this embodiment, each piece of running sub-data is allocated to a corresponding sub-combined external category according to the combined external category, and the amount of data in each category is counted.

[0086] In this embodiment, if the number of category sub-data of a sub-combination external category is 0, the category is considered to be a false sub-combination external category and is removed from the combined external category. At this time, the combined external category will be updated for the first time to remove invalid combinations.

[0087] In this embodiment, for the sub-combination external categories whose number is less than the preset threshold, a merging operation is performed, which is the second update. After the merging, the related category sub-data will also be merged, which is the third update.

[0088] In this embodiment, the category sub-data of all updated combined external categories and all sub-combined external categories are integrated to determine the first operation data for subsequent analysis.

[0089] The beneficial effects of the above technical solution are as follows: obtaining the operating data of multiple solar panels under all external conditions, pre-processing the operating data to determine the combined external categories, and determining the first operating data, which can ensure the quality and validity of the data, improve the efficiency of data processing, and enhance the accuracy of fault analysis and the pertinence of equipment maintenance.

[0090] Embodiment 3:

[0091] An embodiment of the present invention provides a solar panel fault analysis method, which analyzes first operation data to determine a fault type and a fault feature vector of each fault subtype in the fault type, including:

[0092] Based on the category sub-data of each sub-combination external category in the first operation data, determining all operation sub-data marked with faults in the category sub-data as sub-fault data of the corresponding sub-combination external category;

[0093] Extracting features of each piece of fault operation data in the sub-fault data of each sub-combination external category in the combined external category, and determining a sub-fault feature vector of each piece of fault operation data;

[0094] Performing cluster analysis on all fault operation data in the sub-fault data of each sub-combined external category in the combined external category;

[0095] Determine the fault type of each sub-combined external category in the combined external category based on all clusters of the cluster analysis result, wherein the fault type includes multiple fault sub-types;

[0096] Based on all fault operation data in each cluster group of each sub-combination external category, a fault feature vector of a fault sub-type corresponding to each cluster group in each sub-combination external category is determined.

[0097] In this embodiment, operating sub-data marked as faults under each sub-combination external category are extracted from the first operating data, and these fault data are referred to as sub-fault data.

[0098] In this embodiment, cluster analysis is performed on all sub-fault data of each sub-combination external category, and similar fault operation data are classified into one category, so as to identify common features of the faults.

[0099] In this embodiment, based on the result of cluster analysis, the fault data in each sub-combination external category is divided into different fault sub-types, and the fault type of the sub-combination external category is determined according to all fault sub-types of each sub-combination external category.

[0100] In this embodiment, for each cluster group in each sub-combination external category (ie, the category in the clustering result), all fault operation data are further analyzed to determine the fault feature vector of the corresponding fault sub-type.

[0101] In this embodiment, the sub-fault data is the operating data marked as fault in the category sub-data.

[0102] The beneficial effects of the above technical solution are as follows: by analyzing the first operating data, determining the fault type and the fault feature vector of each fault subtype in the fault type, the efficiency of data processing can be further improved, and the accuracy of fault analysis and the pertinence of equipment maintenance can be enhanced.

[0103] Embodiment 4:

[0104] The embodiment of the present invention provides a solar panel fault analysis method, which determines the fault feature vector of the fault subtype corresponding to each cluster group in each subcombination external category based on all fault operation data in each cluster group in each subcombination external category, including:

[0105] Fa i =(f i1 …, f ij , …, f iN1 );

[0106]

[0107] Among them, Fa i represents the fault feature vector of the fault subtype corresponding to the i-th cluster in the cluster analysis results, f i1 、f ij 、f iN1 They represent the feature values ​​of the 1st feature, the jth feature, and the N1th feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, respectively. N1 represents the number of features in the fault feature vector. ij represents the feature fitting value of the jth feature in the fault feature vector of the fault subtype corresponding to the i-th cluster family, Med(X ijk ) represents the feature median of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, Q3 ij 、Q1 ij They represent the third quartile and the first quartile of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, respectively, ij The characteristic skewness of the jth feature in the fault feature vector corresponding to the fault subtype of the i-th cluster family, δ ij The characteristic kurtosis of the jth feature in the fault feature vector corresponding to the fault subtype of the i-th cluster family, σ ij represents the standard deviation of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, C ijrepresents the cluster center feature value of the jth feature of the corresponding fault subtype of the i-th cluster family, X ijk represents the eigenvalue of the jth feature in the sub-fault feature vector of the kth fault operation data in the fault subtype corresponding to the i-th cluster family, iN2 represents the number of fault operation data in the i-th cluster family, μ ij Represents the average value of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family.

[0108] In this embodiment, Represents the relative volatility of the jth feature in the fault feature vector of the corresponding fault type of the i-th cluster.

[0109] In this embodiment, It represents the weight of the feature value of the jth feature in the sub-fault feature vector of the kth fault operation data in the fault subtype corresponding to the i-th cluster family, determined based on the cluster center feature value and average value of the jth feature of the fault subtype corresponding to the i-th cluster family.

[0110] In this embodiment, each fault subtype of each cluster group in each sub-combination external category corresponds to a fault feature vector.

[0111] In this embodiment, the characteristic skewness indicates the symmetry of the distribution of sub-fault characteristic vectors of all data fault operation data in the corresponding fault sub-type.

[0112] In this embodiment, the characteristic kurtosis represents the peak degree of the distribution of the sub-fault characteristic vectors of all data fault operation data in the corresponding fault sub-type.

[0113] The beneficial effects of the above technical solution are as follows: based on all fault operation data in each cluster group of each sub-combination external category, the fault feature vector of the fault subtype corresponding to each cluster group in each sub-combination external category is determined, which can improve the accuracy of fault identification and enhance the understanding of complex fault modes, so as to improve the accuracy of fault analysis and the pertinence of equipment maintenance.

[0114] Embodiment 5:

[0115] The embodiment of the present invention provides a solar panel fault analysis method, which collects real-time operation data of the solar panel in real time, analyzes the real-time operation data to determine a real-time feature vector, and includes:

[0116] Based on the sensor group, real-time operation data of the solar panel is collected in real time, wherein the sensor group includes at least a current sensor, a voltage sensor, a power sensor, a temperature sensor and a radiation intensity sensor, and the real-time operation data includes a plurality of real-time operation sub-data;

[0117] Based on the real-time operation sub-data corresponding to each sensor in the sensor group, feature extraction is performed on each real-time operation sub-data to determine a feature value of each real-time operation sub-data;

[0118] Extracting feature values ​​of all real-time operation sub-data based on external conditions, and determining sub-combination external categories of the real-time operation data based on all extracted feature values ​​and combined external categories;

[0119] A real-time feature vector is determined based on the feature values ​​of all real-time operation sub-data except the external conditions in the real-time operation data.

[0120] In this embodiment, a sensor group is used to collect real-time operating data of the solar panel.

[0121] In this embodiment, the current sensor measures the current output by the solar panel; the voltage sensor measures the voltage output by the solar panel; the power sensor calculates the output power of the solar panel; the temperature sensor monitors the temperature of the solar panel; and the radiation intensity sensor measures the solar radiation intensity (light intensity).

[0122] In this embodiment, the data collected by each sensor is a real-time operation sub-data, and the data collected by all sensors are collectively referred to as real-time operation data.

[0123] In this embodiment, feature extraction is performed on the real-time operation sub-data collected by each sensor, and the feature value reflects the characteristics of the operation state.

[0124] In this embodiment, the feature values ​​of all real-time operation sub-data related to the external conditions are extracted, and all sub-combination external categories in the combination external category are combined to determine to which sub-combination external category the real-time operation data belongs.

[0125] In this embodiment, a real-time feature vector is constructed according to the feature values ​​of all the operation sub-data that are not related to the external conditions in the real-time operation data.

[0126] The beneficial effects of the above technical solution are: real-time collection of real-time operating data of solar panels, analysis of real-time operating data to determine real-time feature vectors, which can ensure the quality and validity of data and improve the efficiency of data processing.

[0127] Embodiment 6:

[0128] An embodiment of the present invention provides a solar panel fault analysis method, which determines the device state of the solar panel based on the real-time feature vector of the solar panel and the fault features of all fault subtypes in the fault types of all sub-combined external categories, including:

[0129] determining similarity of a real-time feature vector of the solar panel and each fault feature vector in a sub-combination external category of the real-time operation sub-data;

[0130] Determine a device state of the solar panel based on all similarity values, wherein the device state includes a faulty device, a device to be faulty, and a normal device;

[0131] If any similarity value is greater than a first fault preset threshold, the solar panel is determined to be a faulty device, and the real-time fault type is determined based on the fault subtypes corresponding to all similarity values ​​greater than the first fault preset threshold;

[0132] If any similarity value is between the first fault preset threshold and the second fault preset threshold, the solar panel is determined to be a device to be faulty, and the real-time fault type to be occurred is determined based on the fault subtypes corresponding to all similarity values ​​between the first fault preset threshold and the second fault preset threshold;

[0133] Otherwise, the solar panel is determined to be a normal device.

[0134] In this embodiment, the similarity between the real-time feature vector of the solar panel and each fault feature vector in the sub-combination external category of the real-time operation sub-data is compared, and the similarity analysis can adopt a variety of methods, such as Euclidean distance, cosine similarity, etc.

[0135] In this embodiment, based on all calculated similarity values, the device status of the solar panel is determined, and the device status can be divided into: faulty device: the similarity between the real-time feature vector and the fault feature vector is higher than the first fault preset threshold, indicating that the solar panel is faulty; faulty device: the similarity value is between the first fault preset threshold and the second fault preset threshold, indicating that the device may be about to fail; normal device: the similarity value is lower than the second fault preset threshold, indicating that the device is operating normally.

[0136] In this embodiment, similarity is a value that quantifies the degree of similarity between two vectors, and is usually calculated using a distance or similarity measurement method.

[0137] In this embodiment, the device status indicates the operating status of the solar cell panel, which is classified into a faulty device, a device about to fail, and a normal device.

[0138] In this embodiment, the first fault preset threshold is a threshold of a similarity value used to determine whether a device has a fault. If the threshold exceeds this value, the device is determined to be a faulty device.

[0139] In this embodiment, the second fault preset threshold is a threshold of the similarity value used to determine whether a device is likely to fail. If the value is between this value and the first threshold, the device is determined to be a device that is about to fail.

[0140] The beneficial effects of the above technical solution are as follows: the device status of the solar panel is determined according to the real-time feature vector of the solar panel and the fault characteristics of all fault subtypes in the fault types of all sub-combined external categories, which can ensure the immediate identification of the fault type, improve the timeliness and accuracy of the fault warning, provide data support for maintenance decisions, effectively reduce the economic losses caused by equipment failures, and improve the overall operation efficiency.

[0141] Embodiment 7:

[0142] An embodiment of the present invention provides a solar panel fault analysis method, which performs fault processing or maintenance on the solar panel based on the device status of the solar panel, including:

[0143] If the solar panel is a faulty device, a fault handling strategy is generated based on the real-time fault type and real-time operating characteristics of the solar panel;

[0144] If the solar panel is a device that is about to fail, a maintenance suggestion is generated based on the real-time type of failure that is about to fail of the solar panel;

[0145] Perform fault handling on solar panels based on fault handling strategies and perform equipment maintenance on solar panels based on maintenance recommendations;

[0146] A fault report is generated based on the real-time fault type, fault handling strategy and fault handling result of the solar panel, and a maintenance report is generated based on the real-time pending fault type, maintenance suggestion and maintenance result of the solar panel.

[0147] In this embodiment, when a solar panel is determined to be a faulty device, a fault handling strategy needs to be formulated based on its real-time fault type and real-time operating characteristics. This strategy includes repair methods, steps and priorities for specific faults to restore the normal operation of the device as soon as possible.

[0148] In this embodiment, if the solar panel is determined to be a device that is about to fail, maintenance suggestions need to be made based on the type of failure that is about to occur. These suggestions may include regular inspections, component replacements, performance optimization, etc., to prevent failures from occurring and extend the service life of the equipment.

[0149] In this embodiment, corresponding fault handling is performed according to the generated fault handling strategy to solve the fault problem existing in the device.

[0150] In this embodiment, equipment maintenance is performed according to maintenance recommendations to prevent potential failures from occurring.

[0151] In this embodiment, based on the real-time fault type of the solar panel, the fault handling strategy and the handling result thereof, a fault report is generated, and the report records the fault condition, handling measures and results in detail.

[0152] In this embodiment, based on the real-time type of fault to be occurred, maintenance suggestions and maintenance results, a maintenance report is generated to record the maintenance process and results, providing a reference for subsequent maintenance and management.

[0153] The beneficial effects of the above technical solution are as follows: based on the equipment status of the solar panel, fault handling or maintenance of the solar panel can improve the fault handling efficiency, ensure preventive maintenance of the equipment in the event of a fault, and significantly improve the operating efficiency and reliability of the equipment.

[0154] Embodiment 8:

[0155] An embodiment of the present invention provides a solar panel fault analysis system, comprising:

[0156] Acquisition module: acquires the operating data of multiple solar panels under all external conditions, pre-processes the operating data to determine the combined external categories, and first operates the data;

[0157] Analysis module: analyzing the first operation data, determining the fault type of each sub-combination external category in the combination external category and the fault feature vector of each fault sub-type in the fault type;

[0158] Acquisition module: collects real-time operation data of solar panels in real time, analyzes the operation data and determines the real-time feature vector;

[0159] Fault module: determines the device status of the solar panel based on the real-time feature vector of the solar panel and the fault features of all fault subtypes in the fault types of all sub-combined external categories;

[0160] Processing module: based on the device status of the solar panel, perform fault processing or maintenance on the solar panel.

[0161] The beneficial effects of the above technical solution are as follows: determining the first operating data through the acquired operating data of the solar panel, analyzing the first operating data to determine the fault type and fault feature vector of the combined external category, determining the device status of the solar panel in combination with the real-time feature vector of the solar panel, and performing processing or maintenance according to the device status, which can improve data processing efficiency, accurately identify the fault type under each sub-combination external category, realize early identification and classification of faults, optimize fault processing and maintenance strategies, significantly improve the operating efficiency and reliability of the equipment, reduce maintenance costs, and extend the service life of the solar panel.

[0162] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0163] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical 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 embodiment or some parts of the embodiments.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A solar panel fault analysis method, characterized in that: include: 101: Acquire operating data of a plurality of solar panels under all external conditions, pre-process the operating data to determine a combined external category, and determine first operating data; 102: Analyze the first operation data to determine the fault type of each sub-combined external category in the combined external category and the fault feature vector of each fault sub-type in the fault type; 103: collecting real-time operation data of the solar panel in real time, and analyzing the real-time operation data to determine a real-time feature vector; 104: Determine a device state of the solar panel based on a real-time feature vector of the solar panel and fault features of all fault subtypes in the fault types of all sub-combined external categories; 105: Based on the device status of the solar panel, perform troubleshooting or maintenance on the solar panel.

2. A solar panel fault analysis method according to claim 1, characterized in that: Acquiring operation data of a plurality of solar panels under all external conditions, preprocessing the operation data to determine a combined external category, and determining first operation data, including: Determine a grouping rule for each external condition, and determine an external classification of the corresponding external condition based on the grouping rule for each external condition; Arrange and combine all external classifications to determine a combined external category, wherein the combined external category includes a plurality of sub-combined external categories; Preprocessing the operation data includes data cleaning, data normalization and data grouping; Allocate each piece of operation sub-data in the operation data after data cleaning and data normalization based on the combined external category, and determine the category sub-data and the number of category sub-data of each sub-combined external category class in the combined external category; Determine that all sub-combination external categories whose category sub-data quantity of the sub-combination external category in the combined external category is 0 are false sub-combination external categories, eliminate all false sub-combination external categories in the combined external category, and perform a first update on the combined external category based on the elimination result; Performing category merging on all sub-combined external categories in the combined external category after the first update, whose category sub-data quantity is less than a preset quantity threshold, and performing data merging on the category sub-data of the sub-combined external categories that are subjected to category merging, performing a second update on the combined external category based on the category merging result, and performing a third update on the category sub-data of each sub-combined external category in the combined external category based on the data merging result; First operating data is determined based on the category sub-data of the combined ambient category after the first update, the second update, and all the sub-combined ambient categories after the third update.

3. A solar panel fault analysis method according to claim 2, characterized in that: Analyzing the first operation data to determine the fault type and the fault feature vector of each fault subtype in the fault type includes: Based on the category sub-data of each sub-combination external category in the first operation data, determining all operation sub-data marked with faults in the category sub-data as sub-fault data of the corresponding sub-combination external category; Extracting features of each piece of fault operation data in the sub-fault data of each sub-combination external category in the combined external category, and determining a sub-fault feature vector of each piece of fault operation data; Performing cluster analysis on all fault operation data in the sub-fault data of each sub-combined external category in the combined external category; Determine the fault type of each sub-combined external category in the combined external category based on all clusters of the cluster analysis result, wherein the fault type includes multiple fault sub-types; Based on all fault operation data in each cluster group of each sub-combination external category, a fault feature vector of a fault sub-type corresponding to each cluster group in each sub-combination external category is determined.

4. A solar panel fault analysis method according to claim 3, characterized in that: Based on all fault operation data in each cluster group of each sub-combination external category, a fault feature vector of a fault sub-type corresponding to each cluster group in each sub-combination external category is determined, including: Fa i =(f i1 ,…,f ij ,…,f iN1 ); Among them, Fa i represents the fault feature vector of the fault subtype corresponding to the i-th cluster in the cluster analysis results, f i1 、f ij 、f iN1 They represent the feature values ​​of the 1st feature, the ith feature, and the N1th feature in the fault feature vector of the fault subtype corresponding to the ith cluster family, respectively. N1 represents the number of features in the fault feature vector. ij represents the feature fitting value of the jth feature in the fault feature vector of the fault subtype corresponding to the i-th cluster family, Med(X ijk ) represents the feature median of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, Q3 ij 、Q1 ij They represent the third quartile and the first quartile of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, Yi j The characteristic skewness of the i-th feature in the fault feature vector corresponding to the fault subtype of the i-th cluster family, δ ij The characteristic kurtosis of the jth feature in the fault feature vector corresponding to the fault subtype of the i-th cluster family, σ ij represents the standard deviation of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family, C ij represents the cluster center feature value of the jth feature of the corresponding fault subtype of the i-th cluster family, X ijk represents the eigenvalue of the i-th feature in the sub-fault feature vector of the k-th fault operation data in the fault subtype corresponding to the i-th cluster family, iN2 represents the number of fault operation data in the i-th cluster family, μ i j represents the average value of the jth feature in the fault feature vector of the corresponding fault subtype of the i-th cluster family.

5. A solar panel fault analysis method according to claim 1, characterized in that: Real-time collection of real-time operating data of solar panels, analysis of real-time operating data to determine real-time feature vectors, including: Based on the sensor group, real-time operation data of the solar panel is collected in real time, wherein the sensor group includes at least a current sensor, a voltage sensor, a power sensor, a temperature sensor and a radiation intensity sensor, and the real-time operation data includes a plurality of real-time operation sub-data; Based on the real-time operation sub-data corresponding to each sensor in the sensor group, feature extraction is performed on each real-time operation sub-data to determine a feature value of each real-time operation sub-data; Extracting feature values ​​of all real-time operation sub-data based on external conditions, and determining sub-combination external categories of the real-time operation data based on all extracted feature values ​​and combined external categories; A real-time feature vector is determined based on the feature values ​​of all real-time operation sub-data except the external conditions in the real-time operation data.

6. A solar panel fault analysis method according to claim 1, characterized in that: Based on the real-time feature vector of the solar panel and the fault features of all fault subtypes in the fault types of all sub-combined external categories, the device status of the solar panel is determined, including: determining similarity of a real-time feature vector of the solar panel and each fault feature vector in a sub-combination external category of the real-time operation sub-data; Determine a device state of the solar panel based on all similarity values, wherein the device state includes a faulty device, a device to be faulty, and a normal device; If any similarity value is greater than a first fault preset threshold, the solar panel is determined to be a faulty device, and the real-time fault type is determined based on the fault subtypes corresponding to all similarity values ​​greater than the first fault preset threshold; If any similarity value is between the first fault preset threshold and the second fault preset threshold, the solar panel is determined to be a device to be faulty, and the real-time fault type to be occurred is determined based on the fault subtypes corresponding to all similarity values ​​between the first fault preset threshold and the second fault preset threshold; Otherwise, the solar panel is determined to be a normal device.

7. A solar panel fault analysis method according to claim 6, characterized in that: Based on the equipment status of the solar panels, troubleshoot or maintain the solar panels, including: If the solar panel is a faulty device, a fault handling strategy is generated based on the real-time fault type and real-time operating characteristics of the solar panel; If the solar panel is a device that is about to fail, a maintenance suggestion is generated based on the real-time type of failure that is about to fail of the solar panel; Perform fault handling on solar panels based on fault handling strategies and perform equipment maintenance on solar panels based on maintenance recommendations; A fault report is generated based on the real-time fault type, fault handling strategy and fault handling result of the solar panel, and a maintenance report is generated based on the real-time pending fault type, maintenance suggestion and maintenance result of the solar panel.

8. A solar panel fault analysis system, characterized in that: include: Acquisition module: acquires the operating data of multiple solar panels under all external conditions, pre-processes the operating data to determine the combined external categories, and first operates the data; Analysis module: analyzing the first operation data, determining the fault type of each sub-combination external category in the combination external category and the fault feature vector of each fault sub-type in the fault type; Acquisition module: collects real-time operation data of solar panels in real time, analyzes the operation data and determines the real-time feature vector; Fault module: determines the device status of the solar panel based on the real-time feature vector of the solar panel and the fault features of all fault subtypes in the fault types of all sub-combined external categories; Processing module: based on the device status of the solar panel, perform fault processing or maintenance on the solar panel.

Images