Fault diagnosis method and device of photovoltaic power station and photovoltaic system

Abstract

The application discloses a photovoltaic power station fault diagnosis method and device and a photovoltaic system, and belongs to the technical field of photovoltaic power generation. The photovoltaic component life prediction method comprises the following steps: determining the corresponding shadow blocking type of the to-be-tested photovoltaic power station based on historical current data of the to-be-tested photovoltaic power station, wherein the shadow blocking type comprises at least one of a fixed blocking, a grass and tree blocking and a whole-day low efficiency; determining the target label of the to-be-tested photovoltaic power station based on at least one of dynamic data and static data of the to-be-tested photovoltaic power station; the target label is used for describing the characteristics of the to-be-tested photovoltaic power station from multiple different dimensions; and determining the target fault information of the to-be-tested photovoltaic power station based on the shadow blocking type and the target label. The photovoltaic power station fault diagnosis method can distinguish the coupling scene and diagnose the specific fault type corresponding to the to-be-tested photovoltaic power station, and can form positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power station by using historical diagnosis results.

Description

Technical Field

[0001] This application belongs to the field of photovoltaic power generation technology, and in particular relates to a fault diagnosis method, device and photovoltaic system for a photovoltaic power station. Background Technology

[0002] Photovoltaic arrays are one of the core components of photovoltaic (PV) power generation systems. Operating in harsh outdoor environments for extended periods, they are prone to anomalies and malfunctions, leading to system failure. Therefore, fault diagnosis of PV arrays is essential for ensuring the normal operation of PV power generation systems. Commonly used fault diagnosis methods for PV power plants have low accuracy. For example, when the shading type of a PV power plant is fixed shading, existing fault diagnosis methods can only identify the fixed shading, but cannot distinguish whether it is shading from a mountain or a large tree, thus limiting their applicability. Summary of the Invention

[0003] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a fault diagnosis method, device, and photovoltaic system for photovoltaic power plants, which can distinguish coupled scenarios and diagnose the specific fault type corresponding to the photovoltaic power plant under test. At the same time, it can use historical diagnosis results to form positive feedback, correction, or guidance for the fault diagnosis of photovoltaic power plants, thereby solving the technical problem of not being able to diagnose specific fault types under coupled scenarios.

[0004] Firstly, this application provides a fault diagnosis method for photovoltaic power plants, the method comprising:

[0005] Based on the historical current data of the photovoltaic power station under test, the shading type of the photovoltaic power station under test is determined. The shading type includes fixed shading, vegetation shading, and inefficiency throughout the day.

[0006] Based on at least one of the dynamic data and static data of the photovoltaic power station under test, a target label is determined for the photovoltaic power station under test; the dynamic data is used to characterize the working status of the photovoltaic power station under test, the static data is used to characterize the static attributes of the photovoltaic power station under test, and the target label is used to describe the characteristics of the photovoltaic power station under test from multiple different dimensions.

[0007] Based on the shadow occlusion type and the target label, the target fault information of the photovoltaic power station under test is determined.

[0008] According to the fault diagnosis method for photovoltaic power plants provided in this application, the shading type of the photovoltaic power plant under test is determined based on the historical current data of the photovoltaic power plant under test. Then, based on at least one of the dynamic and static data of the photovoltaic power plant under test, the target label of the photovoltaic power plant under test is determined. Finally, based on the shading type and the target label, the target fault information of the photovoltaic power plant under test is determined. This method can distinguish the coupled scenarios and diagnose the specific fault type of the photovoltaic power plant under test. At the same time, it can use the historical diagnosis results to form positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power plant, forming a diagnostic closed loop. This improves the precision and accuracy of the final fault diagnosis of the photovoltaic power plant, thereby solving the technical problem of not being able to diagnose the specific fault type in coupled scenarios.

[0009] One embodiment of the fault diagnosis method for a photovoltaic power station according to this application includes determining a target label for the photovoltaic power station under test based on at least one of dynamic data and static data, comprising:

[0010] The target label is determined by describing at least one of the dynamic and static data of the photovoltaic power station under test from at least one dimension.

[0011] The target label includes at least one of fact labels, model labels, and prediction labels; the fact labels are used to describe the factual characteristics of the photovoltaic power station under test, the model labels are determined based on historical diagnostic results, and the prediction labels are used to describe the possible future situations of the photovoltaic power station under test.

[0012] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, by describing at least one of the dynamic and static data of the photovoltaic power plant under test from at least one dimension, the target label is determined. After determining the shading type of the photovoltaic power plant under test, the coupled scenario can be further distinguished based on the target label corresponding to the photovoltaic power plant under test, thus broadening the applicability of the fault diagnosis method for photovoltaic power plants.

[0013] One embodiment of the photovoltaic power station fault diagnosis method of this application, wherein describing at least one of the dynamic data and static data of the photovoltaic power station under test from at least one dimension and determining the target label includes:

[0014] Based on the dynamic data and the static data, the equipment attributes and operating status of the photovoltaic power station under test are determined;

[0015] The device attributes are described quantitatively and / or qualitatively to determine the fact labels;

[0016] The device attributes and operating status are abstracted and clustered to determine the model labels;

[0017] Based on the equipment attributes, the potential faults corresponding to the photovoltaic power station under test are predicted;

[0018] The predicted label is determined based on the potential fault.

[0019] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, by quantitatively and / or qualitatively describing equipment attributes and determining fact labels, it is possible to quantitatively or qualitatively describe the basic attributes or equipment information of the equipment; by abstracting and performing cluster analysis on equipment attributes and operating status, model labels are determined, which can represent the operating conditions or safety status of the equipment; by predicting the potential faults corresponding to the photovoltaic power plant under test based on equipment attributes, and determining the prediction labels based on the potential faults, it is possible to more accurately inspect and eliminate defects in the equipment during practical applications, thereby improving the fault diagnosis accuracy of photovoltaic power plants.

[0020] One embodiment of the photovoltaic power station fault diagnosis method of this application, wherein determining the target fault information of the photovoltaic power station under test based on the shading type and the target label includes:

[0021] Based on the shadow occlusion type, at least one candidate fault information corresponding to the shadow occlusion type is determined;

[0022] Based on the target label, the target fault information is obtained by filtering from the at least one candidate fault information.

[0023] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, by determining at least one candidate fault information corresponding to the shading type based on the shading type, and then filtering the target fault information from the at least one candidate fault information based on the target label, the method can diagnose the specific inefficiency fault type corresponding to the photovoltaic power plant under test. At the same time, it realizes the use of known actual conditions to help decouple the coupled situation itself. It can use historical diagnosis results to form positive feedback, correction or guidance for the fault diagnosis of photovoltaic power plants, forming a diagnosis closed loop, thereby improving the precision and accuracy of fault diagnosis of photovoltaic power plants, and thus solving the technical problem that specific fault types cannot be diagnosed in coupled scenarios.

[0024] One embodiment of the fault diagnosis method for a photovoltaic power station according to this application includes determining the shading type of the photovoltaic power station under test based on historical current data of the photovoltaic power station under test, comprising:

[0025] Based on the historical current data of the photovoltaic power station under test, the inefficient strings in the photovoltaic power station under test are identified.

[0026] The shadow occlusion type is determined based on the sub-historical current data corresponding to the inefficient string.

[0027] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, the inefficient strings in the photovoltaic power plant under test are determined based on the historical current data of the photovoltaic power plant under test, and then the shading type is determined based on the sub-historical current data corresponding to the inefficient strings. This method can effectively diagnose the shading type corresponding to the inefficient strings in the photovoltaic power plant, which facilitates the diagnosis of fault information corresponding to the inefficient strings during subsequent execution.

[0028] A method for fault diagnosis of a photovoltaic power station according to an embodiment of this application, wherein determining the shading type based on the sub-historical current data corresponding to the inefficient string includes:

[0029] Based on the sub-historical current data corresponding to the inefficient string, the target feature corresponding to the sub-historical current data is obtained;

[0030] Based on the target features, the shadow occlusion type is determined.

[0031] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, the target features corresponding to the sub-historical current data corresponding to the inefficient string are obtained based on the sub-historical current data, and the shading type is determined based on the target features. The method can determine the shading type corresponding to the inefficient string corresponding to the feature by analyzing different features of the sub-historical current data. The method has high accuracy and the results obtained are accurate, which can better reflect the operating status of the photovoltaic power plant, thereby helping to improve the accuracy of subsequent fault diagnosis of photovoltaic power plants.

[0032] A method for fault diagnosis of a photovoltaic power station according to an embodiment of this application, wherein determining the shading type based on the sub-historical current data corresponding to the inefficient string includes:

[0033] The sub-historical current data is input into the string inefficiency refined diagnostic model to obtain the shadow occlusion type corresponding to the sub-historical current data output by the string inefficiency refined diagnostic model, wherein...

[0034] The string inefficient refined diagnostic model is trained using sample sub-historical current data as samples and sample shadow occlusion type corresponding to the sample sub-historical current data as sample labels.

[0035] According to the fault diagnosis method for photovoltaic power plants provided in this application, a string inefficiency refined diagnosis model is trained by using sample sub-historical current data as samples and sample shadow occlusion types corresponding to the sample sub-historical current data as sample labels. The sub-historical current data is then input into the string inefficiency refined diagnosis model to obtain the shadow occlusion types corresponding to the sub-historical current data output by the string inefficiency refined diagnosis model. In practical applications, only pre-training is required before use, and then the data can be directly obtained, resulting in high diagnostic efficiency and accuracy. Furthermore, the string inefficiency refined diagnosis model has strong learning capabilities, and the data from each application process can be used as training data for the next training process, thereby improving the model's accuracy and precision. This makes it easier for users to use, broadens the diagnostic scope, and has high versatility, while also improving the accuracy of the final fault diagnosis.

[0036] One embodiment of the present application provides a method for fault diagnosis of a photovoltaic power station, wherein the static data includes at least one of the following: equipment information, defect records, condition-based maintenance records, network security investigation records, inspection records, patrol records, and installed capacity information of the photovoltaic power station under test.

[0037] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, by setting static data including at least one of the following: equipment information, defect records, condition maintenance records, network security investigation records, inspection records, patrol records, and installed capacity information of the photovoltaic power plant under test, the static attributes of the photovoltaic power plant under test can be comprehensively reflected, thereby determining the target label of the photovoltaic power plant under test, and thus enabling more accurate detection of faults in the photovoltaic power plant under test.

[0038] Secondly, this application provides a fault diagnosis device for a photovoltaic power station, the device comprising:

[0039] The first processing module is used to determine the shading type of the photovoltaic power station under test based on the historical current data of the photovoltaic power station under test. The shading type includes at least one of fixed shading, vegetation shading and all-day inefficiency.

[0040] The second processing module is used to determine the target label of the photovoltaic power station under test based on at least one of the dynamic data and static data of the photovoltaic power station under test; the dynamic data is used to characterize the working status of the photovoltaic power station under test, the static data is used to characterize the static attributes of the photovoltaic power station under test, and the target label is used to describe the characteristics of the photovoltaic power station under test from multiple different dimensions.

[0041] The third processing module is used to determine the target fault information of the photovoltaic power station under test based on the shadow occlusion type and the target label.

[0042] According to the photovoltaic power station fault diagnosis device provided in the embodiments of this application, the shading type of the photovoltaic power station under test is determined based on the historical current data of the photovoltaic power station under test. Then, based on at least one of the dynamic and static data of the photovoltaic power station under test, the target label of the photovoltaic power station under test is determined. Based on the shading type and the target label, the target fault information of the photovoltaic power station under test is determined. This device can distinguish the coupled scenario and diagnose the specific fault type of the photovoltaic power station under test. At the same time, it can use the historical diagnosis results to form positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power station, forming a diagnostic closed loop. This improves the precision and accuracy of the final fault diagnosis of the photovoltaic power station, thereby solving the technical problem of not being able to diagnose the specific fault type in the coupled scenario.

[0043] Thirdly, this application provides a photovoltaic system comprising:

[0044] At least one photovoltaic string;

[0045] The fault diagnosis device for photovoltaic power plants as described in the second aspect.

[0046] According to the photovoltaic system provided in the embodiments of this application, by setting at least one photovoltaic string and a fault diagnosis device for the photovoltaic power station in the photovoltaic system, the target fault information corresponding to the photovoltaic power station under test can be effectively diagnosed, thereby solving the technical problem that the specific fault type cannot be diagnosed in the coupled scenario.

[0047] The above-described one or more technical solutions in the embodiments of this application have at least one of the following technical effects:

[0048] By using historical current data of the photovoltaic power station under test, the shading type of the photovoltaic power station under test is determined. Then, based on at least one of the dynamic and static data of the photovoltaic power station under test, the target label of the photovoltaic power station under test is determined. Finally, based on the shading type and the target label, the target fault information of the photovoltaic power station under test is determined. This method can distinguish coupled scenarios and diagnose the specific fault type of the photovoltaic power station under test. At the same time, it can use historical diagnostic results to form positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power station, forming a diagnostic closed loop. This improves the precision and accuracy of the final fault diagnosis of the photovoltaic power station, thereby solving the technical problem of not being able to diagnose specific fault types in coupled scenarios.

[0049] Furthermore, by describing at least one of the dynamic and static data of the photovoltaic power station under test from at least one dimension, and determining the target label, it is possible to further distinguish the coupling scenario based on the target label corresponding to the photovoltaic power station under test after determining the shading type of the photovoltaic power station under test, thus broadening the applicability of the fault diagnosis method for photovoltaic power stations.

[0050] Furthermore, by quantitatively and / or qualitatively describing equipment attributes and determining fact labels, it is possible to quantitatively or qualitatively describe the basic attributes or equipment information of the equipment; by abstracting and clustering equipment attributes and operating status, model labels are determined, which can represent the operating conditions or safety status of the equipment; by predicting potential faults corresponding to the photovoltaic power station under test based on equipment attributes, and determining prediction labels based on potential faults, it is possible to more accurately inspect and eliminate equipment in practical applications, thereby improving the fault diagnosis accuracy of photovoltaic power stations.

[0051] Furthermore, by identifying at least one candidate fault information corresponding to the shadow occlusion type based on the shadow occlusion type, and then filtering the target fault information from the candidate fault information based on the target label, the specific inefficiency fault type of the photovoltaic power station under test can be diagnosed. At the same time, it realizes the use of known actual conditions to help decouple the coupled situation itself. It can use historical diagnostic results to form positive feedback, correction or guidance for the fault diagnosis of photovoltaic power stations, forming a diagnostic closed loop, thereby improving the precision and accuracy of fault diagnosis of photovoltaic power stations, thus solving the technical problem of not being able to diagnose specific fault types in coupled scenarios.

[0052] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0053] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0054] Figure 1 This is one of the flowcharts illustrating the fault diagnosis method for a photovoltaic power station provided in the embodiments of this application;

[0055] Figure 2 This is a schematic diagram illustrating the principle of the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application;

[0056] Figure 3 This is the second flowchart illustrating the fault diagnosis method for a photovoltaic power station provided in the embodiments of this application;

[0057] Figure 4 This is a schematic diagram of the structure of the fault diagnosis device for a photovoltaic power station provided in the embodiments of this application;

[0058] Figure 5 This is a schematic diagram of the structure of the photovoltaic system provided in the embodiments of this application;

[0059] Figure 6This is a schematic diagram of the system architecture of the photovoltaic system provided in the embodiments of this application. Detailed Implementation

[0060] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0061] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0062] The following is combined Figures 1 to 3 This application describes a fault diagnosis method for photovoltaic power plants.

[0063] It should be noted that the entity executing the fault diagnosis method for photovoltaic power plants can be the photovoltaic system, or the fault diagnosis device of the photovoltaic power plant installed on the photovoltaic system, or the server electrically connected to the photovoltaic system, or the user terminal that is communicatively connected to the photovoltaic system, including but not limited to mobile terminals and non-mobile terminals.

[0064] For example, mobile terminals include, but are not limited to, mobile phones, PDA smart terminals, tablets, and in-vehicle smart terminals; non-mobile terminals include, but are not limited to, PCs.

[0065] like Figure 1 As shown, the fault diagnosis method for this photovoltaic power station includes steps 110, 120 and 130.

[0066] Step 110: Based on the historical current data of the photovoltaic power station under test, determine the shading type of the photovoltaic power station under test. The shading types include fixed shading, vegetation shading, and inefficiency throughout the day.

[0067] In this step, the photovoltaic power station under test is used for fault diagnosis.

[0068] Historical current data can be current data of the photovoltaic power station under test prior to the current moment, obtained in chronological order.

[0069] Shadow occlusion types can include at least one of fixed occlusion, vegetation occlusion, and all-day inefficiency.

[0070] Fixed obstructions can include one or more of the following: mountain obstructions, utility pole obstructions, building obstructions, and front and rear row obstructions.

[0071] Vegetation screening can include one or more of the following: vine screening, shrub screening, grass screening, and large tree screening.

[0072] Factors contributing to inefficiency throughout the day may include one or more component factors such as component aging and component breakage, or one or more deposits such as dust, bird droppings, and vines on the components.

[0073] Different types of shadow occlusion exhibit different characteristics in their historical current data. In actual execution, the corresponding shadow occlusion type can be determined based on the differences in the characteristics exhibited.

[0074] In some embodiments, step 110 may include:

[0075] Based on the historical current data of the photovoltaic power station under test, the inefficient strings in the photovoltaic power station under test are identified.

[0076] The type of shadow occlusion is determined based on the sub-historical current data corresponding to the inefficient string.

[0077] In this embodiment, the inefficient string is obtained based on the historical current data of the photovoltaic power station under test. In actual implementation, it can be manifested as the inverter corresponding to the inefficient string having a consistently lower power generation or power generation hours compared to the normal string over several consecutive days.

[0078] Sub-historical current data can be current data prior to the current moment corresponding to an inefficient string obtained in chronological order.

[0079] In actual implementation, based on the historical current data of the photovoltaic power station under test, the inefficient strings in the photovoltaic power station under test are identified, and then the sub-historical current data corresponding to the inefficient strings are obtained. Based on the different characteristics of the sub-historical current data, the shading type corresponding to the inefficient strings can be determined by at least one of the following methods: model method or table lookup method.

[0080] In the lookup table method, it is necessary to establish the correspondence between historical current data and shadow occlusion type in advance, such as by establishing a target table based on historical data, multiple test data or human experience.

[0081] In subsequent applications, the target table can be used to find the shadow occlusion type corresponding to the sub-historical current data.

[0082] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, the inefficient strings in the photovoltaic power plant under test are determined based on the historical current data of the photovoltaic power plant under test, and then the shading type is determined based on the sub-historical current data corresponding to the inefficient strings. This method can effectively diagnose the shading type corresponding to the inefficient strings in the photovoltaic power plant, which facilitates the diagnosis of fault information corresponding to the inefficient strings during subsequent execution.

[0083] In some embodiments, determining the shading type based on the sub-historical current data corresponding to the inefficient string may include:

[0084] Based on the sub-historical current data corresponding to the inefficient string, the target features corresponding to the sub-historical current data are obtained.

[0085] Based on the target features, determine the type of shadow occlusion.

[0086] In this embodiment, the target features can be obtained based on the temporal and spatial performance of the sub-historical current data.

[0087] The target features can be seasonal features, growth features, or shape features.

[0088] In practical applications, when the subhistorical current data changes with the altitude and azimuth of the sun, the target characteristics exhibited by the subhistorical current data are seasonal characteristics, which can determine that the shadow occlusion type is fixed occlusion.

[0089] When the subhistorical current data changes with the growth of vegetation, the target feature exhibited by the subhistorical current data is the growth feature, which can determine that the shading type is vegetation shading.

[0090] When the photovoltaic power station under test cannot receive direct solar radiation, the historical current data shows inefficiency throughout the day. The type of shading can be determined as inefficiency throughout the day based on the shape characteristics of this inefficiency. Figure 2 As shown, this is the inefficient shape of the sub-historical current data throughout the day when there are broken modules in the photovoltaic power station under test. When there are broken modules in the photovoltaic power station under test, the shape characteristics of the corresponding sub-historical current data are similar to those of the sub-historical current data corresponding to normal strings. However, the sub-historical current data values ​​of inefficient strings are lower than those of normal strings throughout the day.

[0091] In actual implementation, the shadow occlusion type can be determined by looking up a table. Based on the target features and target table corresponding to the sub-historical current data, the shadow occlusion type corresponding to the target features can be found, which is the shadow occlusion type corresponding to the inefficient string.

[0092] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, the target features corresponding to the sub-historical current data corresponding to the inefficient string are obtained based on the sub-historical current data, and the shading type is determined based on the target features. The method can determine the shading type corresponding to the inefficient string corresponding to the feature by analyzing different features of the sub-historical current data. The method has high accuracy and the results obtained are accurate, which can better reflect the operating status of the photovoltaic power plant, thereby helping to improve the accuracy of subsequent fault diagnosis of photovoltaic power plants.

[0093] In some embodiments, determining the shading type based on the sub-historical current data corresponding to the inefficient string may further include:

[0094] Input the sub-historical current data into the string inefficiency fine-grained diagnostic model, and obtain the shadow occlusion type corresponding to the sub-historical current data output by the string inefficiency fine-grained diagnostic model.

[0095] In this embodiment, the string inefficiency refined diagnostic model is a pre-trained model, which is trained using sample sub-historical current data as samples and sample shadow occlusion type corresponding to the sample sub-historical current data as sample labels.

[0096] In actual implementation, such as Figure 3 As shown, by inputting the sub-historical current data corresponding to the inefficient string into the string inefficiency fine-grained diagnostic model, the shadow occlusion type corresponding to the sub-historical current data output by the string inefficiency fine-grained diagnostic model can be obtained. For example, the shadow occlusion type can be at least one of fixed occlusion, vegetation occlusion, or all-day inefficiency.

[0097] According to the fault diagnosis method for photovoltaic power plants provided in this application, a string inefficiency refined diagnosis model is trained by using sample sub-historical current data as samples and sample shadow occlusion types corresponding to the sample sub-historical current data as sample labels. The sub-historical current data is then input into the string inefficiency refined diagnosis model to obtain the shadow occlusion types corresponding to the sub-historical current data output by the string inefficiency refined diagnosis model. In practical applications, only pre-training is required before use, and then the data can be directly obtained, resulting in high diagnostic efficiency and accuracy. Furthermore, the string inefficiency refined diagnosis model has strong learning capabilities, and the data from each application process can be used as training data for the next training process, thereby improving the model's accuracy and precision. This makes it easier for users to use, broadens the diagnostic scope, and has high versatility, while also improving the accuracy of the final fault diagnosis.

[0098] Step 120: Based on at least one of the dynamic data and static data of the photovoltaic power station under test, determine the target label of the photovoltaic power station under test; the dynamic data is used to characterize the working status of the photovoltaic power station under test, the static data is used to characterize the static attributes of the photovoltaic power station under test, and the target label is used to describe the characteristics of the photovoltaic power station under test from multiple different dimensions.

[0099] In this step, the target labels are obtained by describing the characteristics of the photovoltaic power station under test from different perspectives, and are used to construct a profile of the photovoltaic power station under test.

[0100] Dynamic data is used to characterize the operating status of the photovoltaic power station under test. Dynamic data may include at least one of the following: current, voltage, active power, reactive power, and signal of the photovoltaic power station under test.

[0101] Static data is used to characterize the static properties of the photovoltaic power plant under test.

[0102] In some embodiments, static data may include at least one of the following: equipment information, defect records, condition-based maintenance records, network security investigation records, inspection records, patrol records, and installed capacity information of the photovoltaic power station under test.

[0103] In some embodiments, the equipment information of the photovoltaic power plant under test may include at least one of the following: model specifications, serial number, asset number, equipment number, manufacturing date, and commissioning date.

[0104] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, the target label of the photovoltaic power plant under test is determined based on at least one of the equipment information, defect records, condition maintenance records, network security investigation records, inspection records, patrol records and installed capacity information of the photovoltaic power plant under test. This method can comprehensively reflect the static attributes of the photovoltaic power plant under test and thus more accurately detect the faults of the photovoltaic power plant under test.

[0105] In some embodiments, step 120 may include: describing at least one of the dynamic data and static data of the photovoltaic power station under test from at least one dimension, and determining the target label;

[0106] The target label includes at least one of fact labels, model labels, and prediction labels; the fact labels are used to describe the factual characteristics of the photovoltaic power station under test, the model labels are determined based on historical diagnostic results, and the prediction labels are used to describe the possible future situations of the photovoltaic power station under test.

[0107] In this embodiment, the target label is used to characterize the data features of the photovoltaic power station under test.

[0108] Each power station can have one or more target tags.

[0109] For example, if the photovoltaic power station to be tested is an industrial or commercial rooftop power station, the target label corresponding to the photovoltaic power station to be tested can be one or more of the following: industrial or commercial power station, dust obstruction, or mountain obstruction.

[0110] In some embodiments, describing at least one of the dynamic and static data of the photovoltaic power station under test from at least one dimension to determine the target label may include:

[0111] Based on dynamic and static data, determine the equipment attributes and operating status of the photovoltaic power station under test;

[0112] Describe equipment attributes quantitatively and / or qualitatively, and determine fact labels;

[0113] Abstracting and clustering equipment attributes and operating status to determine model labels;

[0114] Based on equipment attributes, potential faults corresponding to the photovoltaic power station under test are predicted;

[0115] Based on potential faults, predictive labels are determined.

[0116] In this embodiment, the equipment attributes of the photovoltaic power station under test may include at least one of the basic attributes of the equipment and equipment data.

[0117] The operating status of the photovoltaic power station under test can include at least one of the equipment's operating status and safety status.

[0118] Fact labels are used to describe the factual characteristics of the photovoltaic power station under test. They can be determined by quantitatively describing equipment attributes, qualitatively describing equipment attributes, or by combining both quantitative and qualitative descriptions. The specific method for determining fact labels can be user-defined and is not limited in this application.

[0119] Model labels are determined based on historical diagnostic results. They can be used to abstract and cluster equipment attributes and operating status to determine model labels. Model labels can characterize at least one of the equipment's operating status and safety status.

[0120] In some embodiments, data mining can be performed based on the historical records of the final diagnostic results to form model labels, which can provide positive feedback, correction or guidance to the diagnostic results in subsequent applications, thereby forming a diagnostic closed loop.

[0121] Predictive labels are used to describe the possible future situations of the photovoltaic power station under test. They can predict the potential faults of the photovoltaic power station under test based on at least one of equipment attributes, behaviors, signaling, location and characteristics, and then determine the predictive labels based on the potential faults.

[0122] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, by quantitatively and / or qualitatively describing equipment attributes and determining fact labels, it is possible to quantitatively or qualitatively describe the basic attributes or equipment information of the equipment; by abstracting and performing cluster analysis on equipment attributes and operating status, model labels are determined, which can represent the operating conditions or safety status of the equipment; by predicting the potential faults corresponding to the photovoltaic power plant under test based on equipment attributes, and determining the prediction labels based on the potential faults, it is possible to more accurately inspect and eliminate defects in the equipment during practical applications, thereby improving the fault diagnosis accuracy of photovoltaic power plants.

[0123] In actual implementation, after acquiring the dynamic and static data of the photovoltaic power station under test, behavioral data can be obtained using machine learning methods or data analysis methods. At least one of the following can be statistically analyzed: the inefficiency occurrence time, frequency, and inefficiency type of the inefficient strings in the photovoltaic power station under test. Then, the dynamic and static data of the photovoltaic power station under test can be analyzed from at least one dimension by establishing a behavioral data model to determine the labels of each dimension of the photovoltaic power station under test, which are the target labels corresponding to the photovoltaic power station under test.

[0124] Among them, the photovoltaic power station to be tested can correspond to multiple target labels, and the set of all target labels can be represented as the profile of the photovoltaic power station to be tested.

[0125] In some embodiments, data mining and statistical analysis can be performed based on the historical records of the final diagnostic results to provide feedback and update to the profile, thereby constructing a comprehensive device profile.

[0126] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, by describing at least one of the dynamic and static data of the photovoltaic power plant under test from at least one dimension, the target label is determined. After determining the shading type of the photovoltaic power plant under test, the coupled scenario can be further distinguished based on the target label corresponding to the photovoltaic power plant under test, thus broadening the applicability of the fault diagnosis method for photovoltaic power plants.

[0127] Step 130: Based on the shadow occlusion type and target label, determine the target fault information of the photovoltaic power station under test.

[0128] In this step, such as Figure 3 As shown, the target fault information can be mountain obstruction, utility pole obstruction, vine obstruction, component aging or dust obstruction, etc., and can be determined based on the shadow obstruction type and target label.

[0129] The method for determining target fault information is explained in detail below.

[0130] like Figure 3 As shown, in some embodiments, step 130 may include: determining at least one candidate fault information corresponding to the shadow occlusion type based on the shadow occlusion type;

[0131] Based on the target label, the target fault information is obtained by filtering from at least one candidate fault information.

[0132] In this embodiment, the shading type can be further subdivided into multiple sub-shading types. For example, the sub-shading type can be mountain shading, vine shading, component aging, or dust shading. When there are different sub-shading types in the photovoltaic power station under test, the performance of the sub-historical current data corresponding to the inefficient string in the photovoltaic power station under test may be the same. The candidate fault information is a set of different sub-shading types corresponding to the same performance results.

[0133] In actual execution, after determining at least one candidate fault information corresponding to the shadow occlusion type, the shadow occlusion type can be further subdivided based on the target label to determine the sub-occlusion type, and then the target fault information can be obtained from at least one candidate fault information.

[0134] For example, if the shading type is determined to be inefficient throughout the day, fact labels such as whether the photovoltaic power station under test is an industrial or commercial rooftop power station or whether there is dust shading on the photovoltaic power station under test can be used, or historical diagnostic results such as dust can be statistically analyzed through model labels. This can provide positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power station, and thus determine the target fault information as dust shading.

[0135] In this embodiment, by determining at least one candidate fault information corresponding to the shadow occlusion type based on the shadow occlusion type, and then filtering the target fault information from the at least one candidate fault information based on the target label, the specific inefficiency fault type corresponding to the photovoltaic power station under test can be diagnosed. At the same time, it realizes the use of known actual conditions to help decouple the coupled situation itself. It can use historical diagnostic results to form positive feedback, correction or guidance for the fault diagnosis of photovoltaic power station, forming a diagnostic closed loop, thereby improving the precision and accuracy of fault diagnosis of photovoltaic power station, thus solving the technical problem of not being able to diagnose specific fault types in coupled scenarios.

[0136] In this application, based on the historical current data of the photovoltaic power station under test, the shading type of the photovoltaic power station under test is determined, and the shading type of the photovoltaic power station under test can be preliminarily classified.

[0137] Based on at least one of the dynamic and static data of the photovoltaic power station under test, the target label of the photovoltaic power station under test is determined. The shading type can be further subdivided based on the target label, thereby determining the target fault information of the photovoltaic power station under test.

[0138] According to the fault diagnosis method for photovoltaic power plants provided in the embodiments of this application, the shading type is first determined based on the historical current data of the photovoltaic power plant under test, and then the shading type is further subdivided based on the target label. This allows for the differentiation of coupled scenarios and the diagnosis of the specific fault type corresponding to the photovoltaic power plant under test. At the same time, the historical diagnosis results can be used to form positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power plant, forming a diagnostic closed loop. This improves the precision and accuracy of the final fault diagnosis of the photovoltaic power plant, thereby solving the technical problem of not being able to diagnose the specific fault type in coupled scenarios.

[0139] The fault diagnosis device for photovoltaic power plants provided in this application is described below. The fault diagnosis device for photovoltaic power plants described below can be referred to in correspondence with the fault diagnosis method for photovoltaic power plants described above.

[0140] The fault diagnosis method for photovoltaic power plants provided in this application can be executed by a fault diagnosis device for photovoltaic power plants. This application uses the example of a fault diagnosis device for a photovoltaic power plant executing the fault diagnosis method to illustrate the fault diagnosis device for photovoltaic power plants provided in this application.

[0141] This application also provides a fault diagnosis device for a photovoltaic power station.

[0142] like Figure 4 As shown, the fault diagnosis device for the photovoltaic power station includes: a first processing module 410, a second processing module 420, and a third processing module 430.

[0143] The first processing module 410 is used to determine the shading type of the photovoltaic power station under test based on the historical current data of the photovoltaic power station under test. The shading type includes at least one of fixed shading, vegetation shading and all-day inefficiency.

[0144] The second processing module 420 is used to determine the target label of the photovoltaic power station under test based on at least one of the dynamic data and static data of the photovoltaic power station under test; the dynamic data is used to characterize the working status of the photovoltaic power station under test, the static data is used to characterize the static attributes of the photovoltaic power station under test, and the target label is used to describe the characteristics of the photovoltaic power station under test from multiple different dimensions.

[0145] The third processing module 430 is used to determine the target fault information of the photovoltaic power station under test based on the shadow occlusion type and target label.

[0146] According to the photovoltaic power station fault diagnosis device provided in the embodiments of this application, the shading type of the photovoltaic power station under test is determined based on the historical current data of the photovoltaic power station under test. Then, based on at least one of the dynamic and static data of the photovoltaic power station under test, the target label of the photovoltaic power station under test is determined. Based on the shading type and the target label, the target fault information of the photovoltaic power station under test is determined. This device can distinguish the coupled scenario and diagnose the specific fault type of the photovoltaic power station under test. At the same time, it can use the historical diagnosis results to form positive feedback, correction or guidance for the fault diagnosis of the photovoltaic power station, forming a diagnostic closed loop. This improves the precision and accuracy of the final fault diagnosis of the photovoltaic power station, thereby solving the technical problem of not being able to diagnose the specific fault type in the coupled scenario.

[0147] In some embodiments, the second processing module 420 can also be used to describe at least one of the dynamic data and static data of the photovoltaic power station under test from at least one dimension, and determine the target label;

[0148] The target label includes at least one of fact labels, model labels, and prediction labels; the fact labels are used to describe the factual characteristics of the photovoltaic power station under test, the model labels are determined based on historical diagnostic results, and the prediction labels are used to describe the possible future situations of the photovoltaic power station under test.

[0149] According to the fault diagnosis device for photovoltaic power plants provided in the embodiments of this application, by describing at least one of the dynamic and static data of the photovoltaic power plant under test from at least one dimension, the target label is determined. After determining the shading type of the photovoltaic power plant under test, the coupling scenario can be further distinguished based on the target label of the photovoltaic power plant under test, thus broadening the applicability of the fault diagnosis device for photovoltaic power plants.

[0150] In some embodiments, the fault diagnosis device for the photovoltaic power station may further include a fourth processing module for determining the equipment attributes and operating status of the photovoltaic power station under test based on dynamic data and static data.

[0151] Describe equipment attributes quantitatively and / or qualitatively, and determine fact labels;

[0152] Abstracting and clustering equipment attributes and operating status to determine model labels;

[0153] Based on equipment attributes, potential faults corresponding to the photovoltaic power station under test are predicted;

[0154] Based on potential faults, predictive labels are determined.

[0155] According to the fault diagnosis device for photovoltaic power plants provided in the embodiments of this application, by quantitatively and / or qualitatively describing equipment attributes and determining fact labels, it can quantitatively or qualitatively describe the basic attributes or equipment information of the equipment; by abstracting and clustering the equipment attributes and operating status, it can determine model labels, which can represent the operating conditions or safety status of the equipment; by predicting the potential faults corresponding to the photovoltaic power plant under test based on equipment attributes, and determining the prediction labels based on the potential faults, it can more accurately inspect and eliminate defects in the equipment in practical applications, thereby improving the fault diagnosis accuracy of photovoltaic power plants.

[0156] In some embodiments, the third processing module 430 can also be used to determine at least one candidate fault information corresponding to the shadow occlusion type based on the shadow occlusion type.

[0157] Based on the target label, the target fault information is obtained by filtering from at least one candidate fault information.

[0158] According to the photovoltaic power plant fault diagnosis device provided in the embodiments of this application, by determining at least one candidate fault information corresponding to the shading type based on the shading type, and then filtering the target fault information from the at least one candidate fault information based on the target label, the device can diagnose the specific inefficiency fault type corresponding to the photovoltaic power plant under test. At the same time, it realizes the use of known actual conditions to help decouple the coupled situation itself. It can use historical diagnosis results to form positive feedback, correction or guidance for the fault diagnosis of photovoltaic power plants, forming a diagnosis closed loop, thereby improving the precision and accuracy of fault diagnosis of photovoltaic power plants, and thus solving the technical problem of not being able to diagnose specific fault types in coupled scenarios.

[0159] In some embodiments, the first processing module 410 can also be used to determine the inefficient strings in the photovoltaic power station under test based on the historical current data of the photovoltaic power station under test.

[0160] The type of shadow occlusion is determined based on the sub-historical current data corresponding to the inefficient string.

[0161] According to the photovoltaic power plant fault diagnosis device provided in the embodiments of this application, the inefficient strings in the photovoltaic power plant under test are determined based on the historical current data of the photovoltaic power plant under test, and then the shading type is determined based on the sub-historical current data corresponding to the inefficient strings. This can effectively diagnose the shading type corresponding to the inefficient strings in the photovoltaic power plant, which is convenient for diagnosing the fault information corresponding to the inefficient strings in subsequent execution.

[0162] In some embodiments, the fault diagnosis device for the photovoltaic power station may further include a fifth processing module, used to obtain target features corresponding to the sub-historical current data based on the sub-historical current data corresponding to the inefficient string.

[0163] Based on the target features, determine the type of shadow occlusion.

[0164] According to the photovoltaic power plant fault diagnosis device provided in the embodiments of this application, the target features corresponding to the sub-historical current data are obtained based on the sub-historical current data corresponding to the inefficient string, and the shading type is determined based on the target features. It can determine the shading type corresponding to the inefficient string corresponding to the feature by analyzing different features of the sub-historical current data. The accuracy is high and the results obtained are accurate, which can better reflect the operating status of the photovoltaic power plant, thereby helping to improve the accuracy of subsequent photovoltaic power plant fault diagnosis.

[0165] In some embodiments, the fault diagnosis device for the photovoltaic power station may further include a sixth processing module, used to input sub-historical current data into the string inefficiency refined diagnosis model, and obtain the shading type corresponding to the sub-historical current data output by the string inefficiency refined diagnosis model, wherein...

[0166] The string inefficient refined diagnostic model is trained by using the historical current data of sample sub-data as samples and the sample shadow occlusion type corresponding to the historical current data of sample sub-data as sample labels.

[0167] According to the fault diagnosis device for photovoltaic power plants provided in this application embodiment, a string inefficiency refined diagnosis model is trained by using sample sub-historical current data as samples and sample shadow occlusion types corresponding to the sample sub-historical current data as sample labels. The sub-historical current data is then input into the string inefficiency refined diagnosis model to obtain the shadow occlusion types corresponding to the sub-historical current data output by the string inefficiency refined diagnosis model. In practical applications, only pre-training is required before use, and then data can be directly obtained, resulting in high diagnostic efficiency and accuracy. Furthermore, the string inefficiency refined diagnosis model has strong learning capabilities, and data from each application process can be used as training data for the next training process, thereby improving the model's accuracy and precision. This makes it easier for users to use, broadens the diagnostic scope, and has high versatility, while also improving the accuracy of the final fault diagnosis.

[0168] In some embodiments, the fault diagnosis device for the photovoltaic power station may further include a seventh processing module for static data including at least one of the following: equipment information, defect records, condition-based maintenance records, network security investigation records, inspection records, patrol records, and installation information of the photovoltaic power station under test.

[0169] According to the photovoltaic power station fault diagnosis device provided in the embodiments of this application, by setting static data including at least one of the following: equipment information, defect records, condition maintenance records, network security investigation records, inspection records, patrol records, and installed capacity information of the photovoltaic power station under test, it can comprehensively reflect the static attributes of the photovoltaic power station under test, thereby determining the target label of the photovoltaic power station under test, and thus more accurately detecting the faults of the photovoltaic power station under test.

[0170] The fault diagnosis device for the photovoltaic power station in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit the specific operating system used.

[0171] The fault diagnosis device for photovoltaic power plants provided in this application embodiment can achieve Figures 1 to 3 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.

[0172] This application also provides a photovoltaic system.

[0173] The photovoltaic system includes: at least one photovoltaic string and a fault diagnosis device for a photovoltaic power station as described in any of the above embodiments.

[0174] In this embodiment, the fault diagnosis device for the photovoltaic power station is the fault diagnosis device for the photovoltaic power station described in any of the embodiments above.

[0175] The fault diagnosis device for photovoltaic power plants is used to diagnose faults in photovoltaic strings.

[0176] The fault diagnosis device of the photovoltaic power station is electrically connected to the photovoltaic string to obtain the historical current data of the photovoltaic string, and then to determine the shading type of the photovoltaic power station under test, so as to obtain the target fault information of the photovoltaic power station under test.

[0177] like Figure 5 As shown in the embodiment of this application, a photovoltaic system includes a processor 501, a memory 502, and a computer program stored in the memory 502 and run on the processor 501. When the program is executed by the processor 501, it implements the various processes of the above-described photovoltaic power station fault diagnosis method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0178] In practical applications, the system architecture of a photovoltaic system can be designed based on data and supported by tags, such as... Figure 6 As shown.

[0179] It should be noted that the photovoltaic system in this application embodiment includes the mobile photovoltaic system and the non-mobile photovoltaic system described above.

[0180] According to the photovoltaic system provided in the embodiments of this application, by setting at least one photovoltaic string and a fault diagnosis device for the photovoltaic power station in the photovoltaic system, it can provide system support for photovoltaic array maintenance operations, emergency troubleshooting, and reasonable planning. It can also help decouple the coupled situation based on fact labels, model labels, prediction labels and known actual situation, thereby effectively diagnosing the target fault information corresponding to the photovoltaic power station under test, thus solving the technical problem of not being able to diagnose the specific fault type in coupled scenarios.

[0181] This application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer can execute the fault diagnosis method for photovoltaic power plants provided by the above methods. The method includes: determining the shading type of the photovoltaic power plant under test based on historical current data of the photovoltaic power plant under test; determining the target label of the photovoltaic power plant under test based on at least one of dynamic data and static data of the photovoltaic power plant under test; the dynamic data is used to characterize the working status of the photovoltaic power plant under test, and the static data is used to characterize the static attributes of the photovoltaic power plant under test; and determining the target fault information of the photovoltaic power plant under test based on the shading type and the target label.

[0182] This application also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described photovoltaic power station fault diagnosis method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0183] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0184] This application embodiment also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program is implemented to perform the fault diagnosis method for a photovoltaic power station provided by the methods described above. The method includes: determining the shading type of the photovoltaic power station under test based on historical current data of the photovoltaic power station under test; determining a target label for the photovoltaic power station under test based on at least one of dynamic data and static data of the photovoltaic power station under test; the dynamic data is used to characterize the operating status of the photovoltaic power station under test, and the static data is used to characterize the static attributes of the photovoltaic power station under test; and determining the target fault information of the photovoltaic power station under test based on the shading type and the target label.

[0185] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0186] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0187] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 this application.

Claims

1. A method for diagnosing a fault of a photovoltaic power plant, characterized by, The method comprises the following steps: determining a shadow blocking type corresponding to the to-be-tested photovoltaic power station based on historical current data of the to-be-tested photovoltaic power station, wherein the shadow blocking type comprises a fixed blocking, a grass and tree blocking, and a low efficiency all day; determining a target label of the to-be-tested photovoltaic power station based on at least one of dynamic data and static data of the to-be-tested photovoltaic power station; the dynamic data is used to represent a working condition of the to-be-tested photovoltaic power station, the static data is used to represent a static attribute of the to-be-tested photovoltaic power station, and the target label is used to describe a feature of the to-be-tested photovoltaic power station from multiple different dimensions; determining target fault information of the to-be-tested photovoltaic power station based on the shadow blocking type and the target label; the target label is used to distinguish a coupling scenario based on the target label after it is determined that the shadow blocking type corresponding to the to-be-tested photovoltaic power station is determined; the determining of the target fault information of the to-be-tested photovoltaic power station based on the shadow blocking type and the target label comprises: determining at least one candidate fault information corresponding to the shadow blocking type based on the shadow blocking type, and screening the target fault information from the at least one candidate fault information based on the target label, so as to determine a low efficiency fault type corresponding to the to-be-tested photovoltaic power station.

2. The method of claim 1, wherein the determining of the target label of the to-be-tested photovoltaic power station based on at least one of dynamic data and static data of the to-be-tested photovoltaic power station comprises: describing at least one of the dynamic data and the static data of the to-be-tested photovoltaic power station from at least one dimension to determine the target label; the target label comprises at least one of a fact label, a model label and a prediction label; the fact label is used to describe a factual feature of the to-be-tested photovoltaic power station, the model label is determined based on a historical diagnosis result, and the prediction label is used to describe a possible situation of the to-be-tested photovoltaic power station in the future.

3. The method of claim 2, wherein the determining of the target label by describing at least one of the dynamic data and the static data of the to-be-tested photovoltaic power station from at least one dimension comprises: determining a device attribute and a running state of the to-be-tested photovoltaic power station based on the dynamic data and the static data; quantitatively and / or qualitatively describing the device attribute to determine the fact label; abstractly processing and clustering analyzing the device attribute and the running state to determine the model label; predicting a potential fault corresponding to the to-be-tested photovoltaic power station based on the device attribute; determining the prediction label based on the potential fault.

4. The method of diagnosing a fault of a photovoltaic power plant according to any one of claims 1 to 3, characterized in that, the determining of the target fault information of the to-be-tested photovoltaic power station based on the shadow blocking type and the target label comprises: determining at least one candidate fault information corresponding to the shadow blocking type based on the shadow blocking type; screening the target fault information from the at least one candidate fault information based on the target label.

5. The method of claim 1-3, wherein, the determining of the shadow blocking type corresponding to the to-be-tested photovoltaic power station based on the historical current data of the to-be-tested photovoltaic power station comprises: determining a low efficiency group string in the to-be-tested photovoltaic power station based on the historical current data of the to-be-tested photovoltaic power station. Determine the shadow blocking type based on the sub historical current data corresponding to the low-efficiency string.

6. The method of claim 5, wherein the method further comprises: The determination of the shadow blocking type based on the sub historical current data corresponding to the low-efficiency string includes: Obtain target features corresponding to the sub historical current data based on the sub historical current data corresponding to the low-efficiency string; Determine the shadow blocking type based on the target features.

7. The method of claim 5, wherein the method further comprises: The determination of the shadow blocking type based on the sub historical current data corresponding to the low-efficiency string includes: Input the sub historical current data into a string low-efficiency fine diagnosis model to obtain a shadow blocking type corresponding to the sub historical current data output by the string low-efficiency fine diagnosis model, wherein The string low-efficiency fine diagnosis model is trained by taking sample sub historical current data as samples and taking sample shadow blocking types corresponding to the sample sub historical current data as sample labels.

8. The method of diagnosing a failure of a photovoltaic power plant according to any one of claims 1 to 3, characterized in that, The static data includes at least one of equipment information, defect records, state maintenance records, network security investigation records, inspection records, patrol records, and installed capacity information of the photovoltaic power station to be tested.

9. A fault diagnosis device for a photovoltaic power station, characterized in that, Include: A first processing module configured to determine a shadow blocking type corresponding to a photovoltaic power station to be tested based on historical current data of the photovoltaic power station to be tested, the shadow blocking type including at least one of fixed blocking, grass and wood blocking, and all-day low efficiency; A second processing module configured to determine a target label of the photovoltaic power station to be tested based on at least one of dynamic data and static data of the photovoltaic power station to be tested; The dynamic data is used to represent the working condition of the photovoltaic power station to be tested, the static data is used to represent the static attribute of the photovoltaic power station to be tested, and the target label is used to describe the features of the photovoltaic power station to be tested from multiple different dimensions; A third processing module configured to determine target fault information of the photovoltaic power station to be tested based on the shadow blocking type and the target label; The target label is used to distinguish coupled scenarios based on the target label after determining the shadow blocking type corresponding to the photovoltaic power station to be tested; The third processing module is further configured to determine at least one candidate fault information corresponding to the shadow blocking type based on the shadow blocking type, and filter the target fault information from the at least one candidate fault information based on the target label to determine the low-efficiency fault type corresponding to the photovoltaic power station to be tested.

10. A photovoltaic system characterized by, Include: At least one photovoltaic string; The fault diagnosis device of the photovoltaic power station of claim 9.

Images