CN114785284A - Solar cell fault analysis method and system - Google Patents
Solar cell fault analysis method and system Download PDFInfo
- Publication number
- CN114785284A CN114785284A CN202210700902.0A CN202210700902A CN114785284A CN 114785284 A CN114785284 A CN 114785284A CN 202210700902 A CN202210700902 A CN 202210700902A CN 114785284 A CN114785284 A CN 114785284A
- Authority
- CN
- China
- Prior art keywords
- fault
- target
- battery operation
- label
- template
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 145
- 230000002159 abnormal effect Effects 0.000 claims abstract description 232
- 238000009826 distribution Methods 0.000 claims description 137
- 238000001514 detection method Methods 0.000 claims description 34
- 230000008439 repair process Effects 0.000 claims description 33
- 230000000694 effects Effects 0.000 claims description 31
- 238000003745 diagnosis Methods 0.000 claims description 30
- 238000000034 method Methods 0.000 claims description 21
- 238000005457 optimization Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 230000001364 causal effect Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000011084 recovery Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 4
- 208000024891 symptom Diseases 0.000 claims description 4
- 238000005067 remediation Methods 0.000 claims description 2
- 230000005856 abnormality Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000013461 design Methods 0.000 description 20
- 230000008569 process Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 229910021419 crystalline silicon Inorganic materials 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
- H02S50/00—Monitoring or testing of PV systems, e.g. load balancing or fault identification
- H02S50/10—Testing of PV devices, e.g. of PV modules or single PV cells
- H02S50/15—Testing of PV devices, e.g. of PV modules or single PV cells using optical means, e.g. using electroluminescence
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
Abstract
The embodiment of the application discloses a solar cell fault analysis method and system, wherein fault label characteristics corresponding to a target solar cell system are determined through fault classification confidence degrees of an overall dimension and a detail dimension corresponding to target cell operation abnormal data, so that the problems that the detail characteristics of the target cell operation abnormal data are too much concerned when fault label characteristic analysis is carried out only on the basis of the detail dimension fault classification confidence degrees in the related technology, and actual faults cannot be accurately judged when the fault range of the target solar cell system is large can be solved; the problem that the fault analysis of the target solar cell system is inaccurate due to the fact that the detail features of the abnormal operation data of the target cell cannot be fully utilized when the fault label feature analysis is carried out only based on the integral dimension fault classification confidence coefficient in the related technology can also be solved, and therefore the reliability of the final fault analysis is improved.
Description
Technical Field
The application relates to the technical field of solar cell systems, in particular to a solar cell fault analysis method and system.
Background
A solar cell is a photoelectric semiconductor sheet that directly generates electricity by using sunlight, and can output voltage and generate current under the condition of a loop instantly as long as the solar cell is illuminated under a certain illumination condition, and is physically called a Photovoltaic (abbreviated as PV), which is referred to as Photovoltaic for short. That is, a solar cell is a device that directly converts light energy into electric energy by a photoelectric effect or a photochemical effect. Crystalline silicon solar cells working with photovoltaic effect are the mainstream. In the related art, a solar cell system is a large system combining a plurality of solar cells and a solar cell control module, and a solar cell monitoring system monitors the real-time state of the solar cell system so as to timely acquire the operating state of the solar cell system, particularly monitor whether the solar cell system is in a fault state, and timely repair the fault. However, the current failure analysis method for the solar cell system is not accurate enough.
Disclosure of Invention
The application provides a solar cell fault analysis method and system.
In a first aspect, an embodiment of the present application provides a solar cell fault analysis method, which is executed based on a fault detection server, and includes:
acquiring state flow direction characteristic distribution corresponding to target battery operation abnormal data uploaded by a solar battery monitoring system; the target battery abnormal operation data is data with abnormal operation activities obtained by monitoring state data of a target solar battery system; the target solar cell system is a cell target to be subjected to fault analysis;
performing member activity positioning on the state flow direction characteristic distribution, and outputting first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data;
loading the first member state flow direction characteristics into respective associated first target member fault analysis models in a distributed manner, and outputting a first fault classification confidence degree corresponding to the target battery operation abnormal data; the first fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature;
loading the state flow direction characteristic distribution to a target overall fault analysis model, and outputting a second fault classification confidence corresponding to the target battery operation abnormal data; the second fault classification confidence level comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature;
and determining the corresponding fault label characteristics of the target solar cell system based on the first fault classification confidence and the second fault classification confidence.
In a second aspect, an embodiment of the present application provides a fault detection server, including:
a processor;
a memory having stored therein a computer program that, when executed, implements the solar cell failure analysis method of the first aspect.
As described above, the state flow direction characteristic distribution corresponding to the target battery operation abnormal data is obtained; the target battery operation abnormal data is data which is obtained by monitoring state data of a target solar battery system needing fault analysis and has abnormal operation activities; then, performing member activity positioning on the state flow direction characteristic distribution, and outputting a first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data; loading the flow direction characteristics of each first member state into the fault analysis model of each associated first target member, and outputting a first fault classification confidence corresponding to the abnormal operation data of the target battery; the first fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature; meanwhile, loading the state flow direction characteristic distribution to a target overall fault analysis model, and outputting a second fault classification confidence coefficient corresponding to the target battery operation abnormal data; the second fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic; determining a fault label characteristic corresponding to the target solar cell system based on the first fault classification confidence and the second fault classification confidence; therefore, the first member state flow direction characteristics corresponding to each battery operation abnormal activity in the target battery operation abnormal data are loaded into the first target member fault analysis models associated with the first member state flow direction characteristics, so that each battery operation abnormal activity in the target battery operation abnormal data has one corresponding first target member fault analysis model for fault label characteristic analysis, and therefore fault label characteristic analysis can be effectively carried out on a target solar battery system in the target battery operation abnormal data on the basis of the refined member characteristics of the target battery operation abnormal data, and a first fault classification confidence coefficient corresponding to the target battery operation abnormal data is output; by directly loading the state flow direction characteristic distribution to the target integral fault analysis model, the integral characteristic of the target battery operation abnormal data can be fully utilized to carry out fault label characteristic analysis on the target solar battery system, and a second fault classification confidence coefficient corresponding to the target battery operation abnormal data is output; finally, determining the corresponding fault label characteristics of the target solar cell system through the first fault classification confidence coefficient and the second fault classification confidence coefficient corresponding to the abnormal operation data of the target cell, so that the problem that the detailed characteristics of the abnormal operation data of the target cell are too much concerned when fault label characteristic analysis is carried out only based on the first fault classification confidence coefficient in the related technology, and the actual fault cannot be accurately judged when the fault range of the target solar cell system is large can be solved; the problem that the fault analysis of the target solar cell system is inaccurate due to the fact that the detail features of the abnormal operation data of the target cell cannot be fully utilized when the fault label feature analysis is carried out only based on the second fault classification confidence coefficient in the related technology can also be solved, and therefore the reliability of the final fault analysis is improved.
Drawings
Fig. 1 is a schematic flow chart illustrating steps of a method for analyzing a solar cell fault according to an embodiment of the present disclosure;
fig. 2 is a block diagram schematically illustrating a structure of a fault detection server for executing the solar cell fault analysis method in fig. 1 according to an embodiment of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by a person of ordinary skill in the art without making any creative effort based on the embodiments in the present application belong to the protection scope of the present application.
The following describes a specific embodiment of an example of the present application with reference to fig. 1.
STEP110, obtaining the state flow direction characteristic distribution corresponding to the target battery operation abnormal data.
The target battery abnormal operation data is data with abnormal operation activities obtained by monitoring state data of the target solar battery system.
The target solar cell system is a cell target to be subjected to fault analysis.
In some exemplary design ideas, a fault detection server may acquire target battery operation abnormal data obtained by monitoring state data of a target solar battery system which needs to be subjected to fault analysis, load the target battery operation abnormal data into a target deep learning network which meets a model convergence requirement, and perform feature coding on the target battery operation abnormal data through a feature extraction function by a coding unit in the target deep learning network to generate state flow direction feature distribution.
For example, the feature extraction function may be F, if the input target battery operation abnormal data is W, the state flow direction feature distribution generated by the encoding unit is E, and the calculation formula is as follows: e = f (w).
And STEP120, performing member activity positioning on the state flow direction characteristic distribution, and outputting first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data.
In some exemplary design concepts, the fault detection server may perform member activity positioning on the state flow direction characteristic distribution, and output the member activity positioning to the first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data.
STEP130, loading the flow direction characteristics of each first member state into the first target member fault analysis model associated with each first member state, and outputting a first fault classification confidence corresponding to the target battery operation abnormal data.
And the first fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic.
The fault detection server loads each first member state flow direction feature distribution into a first target member fault analysis model associated with each first member state flow direction feature distribution, and a first fault classification confidence coefficient corresponding to target battery operation abnormal data can be obtained through a fault classification confidence coefficient corresponding to each first member state flow direction feature distribution generated by each first target member fault analysis model, wherein the first fault classification confidence coefficient comprises support degrees of each battery operation abnormal activity in the target battery operation abnormal data to each fault label feature.
STEP140, loading the state flow direction characteristic distribution to a target overall fault analysis model, and outputting a second fault classification confidence corresponding to the target battery operation abnormal data.
And the second fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic.
In some exemplary design ideas, the fault detection server may directly load the state flow direction feature distribution to a globally shared target overall fault analysis model, and output a fault classification confidence generated by the target overall fault analysis model for the state flow direction feature distribution, so that a second fault classification confidence corresponding to the target battery operation abnormal data may be obtained based on the fault classification confidence, where the second fault classification confidence includes a support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature.
STEP150, determining a corresponding fault label feature of the target solar cell system based on the first fault classification confidence and the second fault classification confidence.
In some exemplary design considerations, the fault detection server may determine a target fault classification confidence corresponding to the target battery operation abnormal data based on the first fault classification confidence and the second fault classification confidence, where the target fault classification confidence includes a target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault tag feature, and thus the fault detection server may generate the fault tag feature corresponding to the target solar cell system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault tag feature.
For example, the corresponding fault signature of the target solar cell system may be a fault signature of the target solar cell system.
By adopting the technical scheme, the state flow direction characteristic distribution corresponding to the abnormal operation data of the target battery is obtained; the target battery operation abnormal data is data which is obtained by monitoring state data of a target solar battery system needing fault analysis and has abnormal operation activities; then, performing member activity positioning on the state flow direction characteristic distribution, and outputting a first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data; loading the flow direction characteristics of each first member state into a first target member fault analysis model associated with each first member state, and outputting a first fault classification confidence corresponding to the target battery operation abnormal data; the first fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature; meanwhile, loading the state flow direction characteristic distribution to a target overall fault analysis model, and outputting a second fault classification confidence degree corresponding to the target battery operation abnormal data; the second fault classification confidence coefficient comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label characteristic; determining a fault label characteristic corresponding to the target solar cell system based on the first fault classification confidence and the second fault classification confidence; therefore, the first member state flow direction characteristics corresponding to each battery operation abnormal activity in the target battery operation abnormal data are loaded into the first target member fault analysis models associated with the first member state flow direction characteristics, so that each battery operation abnormal activity in the target battery operation abnormal data has one corresponding first target member fault analysis model for fault label characteristic analysis, and therefore fault label characteristic analysis can be effectively carried out on a target solar battery system in the target battery operation abnormal data on the basis of the refined member characteristics of the target battery operation abnormal data, and a first fault classification confidence coefficient corresponding to the target battery operation abnormal data is output; by directly loading the state flow direction characteristic distribution to the target integral fault analysis model, the integral characteristic of the target battery operation abnormal data can be fully utilized to carry out fault label characteristic analysis on the target solar battery system, and a second fault classification confidence coefficient corresponding to the target battery operation abnormal data is output; finally, determining the fault label characteristics corresponding to the target solar cell system through the first fault classification confidence coefficient and the second fault classification confidence coefficient corresponding to the target battery operation abnormal data, so that the problem that the detailed characteristics of the target battery operation abnormal data are too much concerned when fault label characteristic analysis is carried out only on the basis of the first fault classification confidence coefficient in the related technology and the actual fault cannot be accurately judged when the fault range of the target solar cell system is large can be solved; the problem that the fault analysis of the target solar cell system is inaccurate due to the fact that the detail features of the abnormal operation data of the target cell cannot be fully utilized when the fault label feature analysis is carried out only based on the second fault classification confidence coefficient in the related technology can also be solved, and therefore the reliability of the final fault analysis is improved.
In some exemplary design concepts, STEP130 in the above embodiments includes:
STEP210, loading each first member state flow direction feature distribution into the first target member fault analysis model associated with each first member state flow direction feature distribution, and outputting a fault classification confidence corresponding to each first member state flow direction feature distribution.
In some exemplary design ideas, in a process of loading, by the fault detection server, each first member state flow direction feature distribution into a respective associated first target member fault analysis model and outputting a first fault classification confidence corresponding to the target battery operation abnormal data, the fault detection server may load each first member state flow direction feature distribution into a respective associated first target member fault analysis model, and first obtain a fault classification confidence corresponding to each first member state flow direction feature distribution generated by each first target member fault analysis model.
STEP220, fusing the flow direction characteristic distribution of each first member state according to the time-space domain information of the flow direction characteristic distribution of each first member state, and outputting the fused flow direction characteristic distribution of the first member state.
In some exemplary design ideas, after the fault detection server obtains the fault classification confidence corresponding to each first member state flow direction feature distribution generated by each first target member fault analysis model, the fault detection server may fuse each first member state flow direction feature distribution according to time-space domain information of each first member state flow direction feature distribution, and output the fused first member state flow direction feature distribution.
STEP230, based on the fault classification confidence corresponding to each fused first member state flow direction feature distribution, generates a fault classification confidence corresponding to the state flow direction feature distribution.
And the fault classification confidence corresponding to the state flow characteristic distribution comprises the support degree of each battery operation abnormal activity in the state flow characteristic distribution to each fault label characteristic.
In some exemplary design ideas, after obtaining the fused first member state flow direction feature distribution, the fault detection server may generate a fault classification confidence corresponding to the state flow direction feature distribution based on a fault classification confidence corresponding to each fused first member state flow direction feature distribution, where the fault classification confidence corresponding to the state flow direction feature distribution includes support degrees of each battery operation abnormal activity in the state flow direction feature distribution for each fault label feature.
STEP240 generates a first fault classification confidence level based on the fault classification confidence level corresponding to the state flow characteristic distribution.
In some exemplary design ideas, since the state flow direction feature distribution generated by the target deep learning network is consistent with the feature distribution range of the input abnormal battery operation data (i.e., the target abnormal battery operation data), after obtaining the fault classification confidence corresponding to the state flow direction feature distribution, the fault detection server may generate the support degree of each abnormal battery operation activity in the target abnormal battery operation data for each fault label feature based on the support degree of each abnormal battery operation activity in the state flow direction feature distribution for each fault label feature, and output a first fault classification confidence.
By adopting the steps, loading the flow direction characteristic distribution of each first member state into the fault analysis model of the first target member associated with each first member state, and outputting the fault classification confidence corresponding to the flow direction characteristic distribution of each first member state; fusing the flow direction characteristic distribution of each first member state according to the time-space domain information of the flow direction characteristic distribution of the first member state, and outputting the fused flow direction characteristic distribution of the first member state; generating a fault classification confidence coefficient corresponding to the state flow direction feature distribution based on the fault classification confidence coefficient corresponding to the fused first member state flow direction feature distribution; the fault classification confidence corresponding to the state flow direction feature distribution comprises the support degree of each battery operation abnormal activity in the state flow direction feature distribution to each fault label feature; finally, generating a first fault classification confidence coefficient based on the fault classification confidence coefficient corresponding to the state flow direction feature distribution; therefore, the first member state flow direction characteristics corresponding to each battery operation abnormal activity in the target battery operation abnormal data are loaded to the first target member fault analysis models associated with the first member state flow direction characteristics in a distributed manner, so that each battery operation abnormal activity in the target battery operation abnormal data has one first target member fault analysis model corresponding to each battery operation abnormal activity to perform fault label characteristic analysis, therefore, the fault label characteristic analysis can be effectively performed on the target solar battery system in the target battery operation abnormal data on the basis of the refined member characteristics of the target battery operation abnormal data, and the local detail characteristic capture capability in the fault classification process of the target solar battery system is improved.
In some exemplary design considerations, determining a fault label characteristic corresponding to the target solar cell system based on the first fault classification confidence level and the second fault classification confidence level includes: adding the first fault classification confidence coefficient and the second fault classification confidence coefficient, and outputting a target fault classification confidence coefficient corresponding to the target battery operation abnormal data; and generating a fault label characteristic corresponding to the target solar cell system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic.
And the target fault classification confidence degree comprises a target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic.
In some exemplary design ideas, in the process of determining, by the fault detection server, the fault tag feature corresponding to the target solar cell system based on the first fault classification confidence level and the second fault classification confidence level, the fault detection server may add the first fault classification confidence level and the second fault classification confidence level, and output a target fault classification confidence level corresponding to the target abnormal battery operation data, where the target fault classification confidence level includes a target support degree of each abnormal battery operation activity in the target abnormal battery operation data for each fault tag feature; then, based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature, the fault label feature corresponding to the target solar battery system can be determined.
By adopting the steps, the first fault classification confidence coefficient and the second fault classification confidence coefficient are added, and the target fault classification confidence coefficient corresponding to the target battery operation abnormal data is output; the target fault classification confidence comprises a target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic; generating a fault label characteristic corresponding to the target solar cell system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic; therefore, the problem that when fault label feature analysis is carried out only based on the first fault classification confidence coefficient in the related technology, detailed features of target battery operation abnormal data are too much concerned, and when the fault range of a target solar battery system is large, actual faults cannot be accurately judged can be solved; the problem that the fault analysis of the target solar cell system is inaccurate due to the fact that the detail features of the abnormal operation data of the target cell cannot be fully utilized when the fault label feature analysis is carried out only based on the second fault classification confidence coefficient in the related technology can also be solved, and therefore the reliability of the final fault analysis is improved.
In some exemplary design ideas, generating a fault label feature corresponding to a target solar cell system based on a target support degree of each battery operation abnormal activity in target battery operation abnormal data for each fault label feature includes: generating a maximum target support degree corresponding to each battery operation abnormal activity in the target battery operation abnormal data in the target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic; outputting the fault label characteristics corresponding to the maximum target support degrees as the fault label characteristics corresponding to the abnormal operation activities of the batteries in the abnormal operation data of the target batteries; and generating a fault label characteristic corresponding to the target solar cell system based on the fault label characteristic corresponding to each battery operation abnormal activity in the target battery operation abnormal data.
In some exemplary design ideas, in the process of generating the fault tag feature corresponding to the target solar cell system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault tag feature, the fault detection server may generate the maximum target support degree corresponding to each battery operation abnormal activity in the target battery operation abnormal data in the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault tag feature, so that the fault tag feature corresponding to each maximum target support degree may be output as the fault tag feature corresponding to each battery operation abnormal activity in the target battery operation abnormal data, and based on the fault tag feature corresponding to each battery operation abnormal activity in the target battery operation abnormal data, the pair of each battery operation abnormal activity constituting the target solar cell system in the target battery operation abnormal data may be determined And determining the fault label characteristics of the target solar cell system according to the corresponding fault label characteristics.
By adopting the steps, the maximum target support degree corresponding to each battery operation abnormal activity in the target battery operation abnormal data is generated in the target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic; outputting the fault label characteristics corresponding to the maximum target support degrees as the fault label characteristics corresponding to the abnormal operation activities of the batteries in the abnormal operation data of the target batteries; generating fault label characteristics corresponding to the target solar cell system based on the fault label characteristics corresponding to the abnormal operation activities of the batteries in the abnormal operation data of the target battery; therefore, the fault label characteristics to which the maximum target support degree corresponding to each battery operation abnormal activity in the target battery operation abnormal data belongs are output as the fault label characteristics corresponding to each battery operation abnormal activity in the target battery operation abnormal data; therefore, the fault label characteristics corresponding to the target solar cell system can be accurately determined based on the fault label characteristics corresponding to the abnormal operation activities of the cells forming the target solar cell system, and the reliability of fault analysis of the target solar cell system is improved.
In some exemplary design concepts, the method further comprises: performing member activity positioning on the state flow direction characteristic distribution for multiple times under different characteristic dimensions, and outputting second member state flow direction characteristic distribution corresponding to the target battery operation abnormal data under at least one characteristic dimension; loading the flow direction characteristics of each second member state into a respective associated second target member fault analysis model in a distributed manner, and outputting at least one third fault classification confidence corresponding to the target battery operation abnormal data; the at least one third fault classification confidence coefficient comprises the support degree of each fault label characteristic under the corresponding characteristic dimension of each battery operation abnormal activity in the target battery operation abnormal data; and adding the first fault classification confidence coefficient, the second fault classification confidence coefficient and at least one third fault classification confidence coefficient to output a target fault classification confidence coefficient.
And the distribution range of the second member state flow direction characteristic distribution is larger than that of the first member state flow direction characteristic distribution and smaller than that of the state flow direction characteristic distribution.
In some exemplary design considerations, in the process of determining the target fault classification confidence corresponding to the target battery operation abnormal data, the fault detection server may further perform member activity positioning on the state flow direction characteristic distribution for multiple times under different characteristic dimensions, output a second member state flow direction characteristic distribution corresponding to the target battery operation abnormal data under at least one characteristic dimension, and a distribution range of the second member state flow direction characteristic distribution is greater than a distribution range of the first member state flow direction characteristic distribution and smaller than a distribution range of the state flow direction characteristic distribution.
Then, loading the characteristic distribution of the state flow direction of each second member into a second target member fault analysis model associated with each second member, and outputting fault classification confidence degrees generated by the second target member fault analysis models corresponding to different characteristic dimensions, so as to obtain at least one third fault classification confidence degree corresponding to the abnormal operation data of the target battery, wherein the at least one third fault classification confidence degree comprises the support degree of each abnormal operation activity of each battery in the abnormal operation data of the target battery in the corresponding characteristic dimension aiming at each fault label characteristic; and finally, adding the first fault classification confidence coefficient and the second fault classification confidence coefficient with corresponding third fault classification confidence coefficients under different feature dimensions by the fault detection server to obtain a target fault classification confidence coefficient corresponding to the target battery operation abnormal data.
By adopting the steps, member activity positioning under different characteristic dimensions is carried out on the state flow direction characteristic distribution for multiple times, and second member state flow direction characteristic distribution corresponding to the target battery operation abnormal data under at least one characteristic dimension is output; the distribution range of the second member state flow direction characteristic distribution is larger than that of the first member state flow direction characteristic distribution and smaller than that of the state flow direction characteristic distribution; loading the flow direction characteristics of each second member state into the fault analysis model of each associated second target member, and outputting at least one third fault classification confidence corresponding to the abnormal operation data of the target battery; the at least one third fault classification confidence coefficient comprises the support degree of each fault label feature under the corresponding feature dimension of each battery operation abnormal activity in the target battery operation abnormal data; adding the first fault classification confidence coefficient, the second fault classification confidence coefficient and at least one third fault classification confidence coefficient, and outputting a target fault classification confidence coefficient; thus, by carrying out multi-feature dimension division on the state flow direction feature distribution, loading the divided member state flow direction feature distribution into respective associated second target member fault analysis models, obtaining third fault classification confidence coefficients corresponding to the state flow direction feature distribution under different feature dimensions through the output result of each second target member fault analysis model, adding the third fault classification confidence coefficients with the first fault classification confidence coefficients fully utilizing the state flow direction feature distribution detail features and the second fault classification confidence coefficients fully utilizing the state flow direction feature distribution overall features, and outputting the target fault classification confidence coefficients; therefore, semantic content of the fault analysis model can be changed based on the change of state flow direction characteristic distribution content of different regions, characteristic information of state flow direction characteristic distribution in different distribution ranges can be fully extracted and fused, characteristic details of a target solar cell system in the abnormal operation data of the target cell are enriched, the fault label characteristics of the target solar cell system can be accurately identified through the target fault classification confidence corresponding to the abnormal operation data of the target cell, and the self-adaptive perception capability of the fault analysis model to different regions of the abnormal operation data of each target cell is enhanced.
In some exemplary designs, the method of the above embodiment further includes: acquiring a member fault analysis model waiting for model weight optimization; performing model weight information optimization on the member fault analysis model waiting for model weight optimization based on a gradient descent algorithm, and outputting a first member fault analysis model meeting the model convergence requirement; and/or performing model weight information optimization on the member fault analysis model waiting for model weight optimization based on a specified loss function, and outputting a second member fault analysis model meeting the model convergence requirement; and outputting a target member fault analysis model based on the first member fault analysis model and/or the second member fault analysis model.
The member fault analysis models waiting for model weight optimization comprise unit fault analysis models corresponding to different characteristic dimensions.
The target member fault analysis model comprises a first target member fault analysis model, a target overall fault analysis model and a second target member fault analysis model.
In some exemplary design ideas, the fault detection server may further obtain corresponding member fault analysis models waiting for model weight optimization with state flow direction characteristics distributed under different feature dimensions, perform model weight information optimization on the member fault analysis models waiting for model weight optimization based on a gradient descent algorithm, and output a first member fault analysis model meeting a model convergence requirement, where the first member fault analysis model includes d-dimensional fault association characteristic representations corresponding to each fault label characteristic.
Finally, the fault detection server can output a target member fault analysis model based on the first member fault analysis model and/or the second member fault analysis model; the target member fault analysis model comprises a first target member fault analysis model, a target overall fault analysis model and a second target member fault analysis model.
By adopting the steps, the member fault analysis model waiting for model weight optimization is obtained by correspondingly distributing state flow direction characteristics under different characteristic dimensions; performing model weight information optimization on the member fault analysis model waiting for model weight optimization based on a gradient descent algorithm, and outputting a first member fault analysis model meeting the model convergence requirement; and/or performing model weight information optimization on the member fault analysis model waiting for model weight optimization based on a specified loss function, and outputting a second member fault analysis model meeting the model convergence requirement; outputting a target member fault analysis model based on the first member fault analysis model and/or the second member fault analysis model; therefore, the target member fault analysis model can be obtained through training by various methods, the advantages of the various methods are combined, the accuracy of the target member fault analysis model on the fault label characteristic analysis can be improved, and meanwhile, the method for obtaining the target member fault analysis model is more diversified.
In some exemplary design ideas, the target overall fault analysis model comprises a fault comparison unit and a fault classification unit; loading the state flow direction characteristic distribution to a target overall fault analysis model, and outputting a second fault classification confidence corresponding to the target battery operation abnormal data, wherein the second fault classification confidence comprises the following steps: loading the state flow direction characteristics to a fault comparison unit in a distributed manner, and outputting target association parameter values between fault association characteristics corresponding to abnormal operation activities of each battery and fault association characteristics corresponding to each fault label characteristic in target battery operation abnormal data; and loading the target associated parameter values to a fault classification unit, and outputting a second fault classification confidence coefficient.
And the dimension of the fault association characteristic corresponding to each battery operation abnormal activity in the target battery operation abnormal data is equal to the dimension of the fault association characteristic corresponding to each fault label characteristic.
And the second fault classification confidence coefficient is obtained by loading the target associated parameter value to the normalized multi-classification function through the fault classification unit.
In some exemplary design concepts, the target overall fault analysis model comprises a fault comparison unit and a fault classification unit; in the process that the fault detection server loads the state flow direction characteristic distribution to the target overall fault analysis model and outputs the second fault classification confidence degree corresponding to the target battery operation abnormal data, the fault detection server can load the state flow direction characteristic distribution to the fault comparison unit and output target association parameter values between fault association characteristics corresponding to each battery operation abnormal activity in the target battery operation abnormal data and fault association characteristics corresponding to each fault label characteristic; the dimension of the fault association characteristic corresponding to each battery operation abnormal activity in the target battery operation abnormal data is equal to the dimension of the fault association characteristic corresponding to each fault label characteristic; loading the target associated parameter values to a fault classification unit, and outputting a second fault classification confidence coefficient; and the second fault classification confidence coefficient is obtained by loading the target associated parameter value to the normalized multi-classification function through the fault classification unit.
By adopting the steps, the state flow direction characteristic distribution is loaded to the fault comparison unit, and target correlation parameter values between the fault correlation characteristics corresponding to the abnormal operation activities of each battery and the fault correlation characteristics corresponding to the fault label characteristics in the target battery operation abnormal data are output; loading the target associated parameter values to a fault classification unit, and outputting a second fault classification confidence coefficient; therefore, when the fault range of the target solar cell system is large, the target overall fault analysis model can be used for carrying out comprehensive fault classification, and the fault label characteristics corresponding to the target solar cell system can be accurately identified through the generated second fault classification confidence coefficient.
In some exemplary designs, another embodiment may specifically include the following steps:
STEP302, obtaining the state flow direction characteristic distribution corresponding to the target battery operation abnormal data.
STEP304, performing member activity positioning on the state flow direction characteristic distribution, and outputting a first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data.
STEP306, loading each first member state flow direction feature distribution into the first target member fault analysis model associated with each first member state flow direction feature distribution, and outputting a fault classification confidence corresponding to each first member state flow direction feature distribution.
STEP308, merging the flow direction characteristic distribution of each first member state according to the time-space domain information of the flow direction characteristic distribution of each first member state, and outputting the merged flow direction characteristic distribution of the first member state.
STEP310, based on the fault classification confidence corresponding to each fused first member state flow direction feature distribution, generates a fault classification confidence corresponding to the state flow direction feature distribution.
STEP312, based on the fault classification confidence corresponding to the state flow characteristic distribution, generates a first fault classification confidence corresponding to the target battery operation abnormal data.
STEP314, loading the state flow direction characteristics to the fault comparison unit, and outputting a target association parameter value between the fault association characteristic corresponding to each battery operation abnormal activity in the target battery operation abnormal data and the fault association characteristic corresponding to each fault label characteristic.
STEP316, loading the target associated parameter value to the fault classification unit, and outputting a second fault classification confidence corresponding to the target battery operation abnormal data.
STEP318, adding the first fault classification confidence coefficient and the second fault classification confidence coefficient, and outputting a target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature.
STEP320, generating a fault label feature corresponding to the target solar battery system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature.
On the basis of the above description, in a further fault repair phase, the embodiments of the present application may further include the steps of the following embodiments.
STEP160, performing fault repairing on the target solar cell system according to the fault label characteristic corresponding to the target solar cell system.
STEP160 can be implemented by the following substeps.
STEP161, inputting the fault label feature corresponding to the target solar battery system into a fault repair scheme recommendation model, and outputting fault repair scheme information corresponding to the target solar battery system;
STEP162, performing fault repairing on the target solar cell system based on the fault repairing scheme information.
For example, the aforementioned training step of the fault remediation scheme recommendation model includes:
STEP1601, extracting template fault label features from a fault repair template database, and template fault repair scheme information corresponding to each corresponding fault label node in the template fault label features;
STEP1602, combining a failure causal relationship graph extraction unit of a failure repair scheme recommendation model, performing failure causal relationship graph extraction on the template failure label features, and outputting template failure causal relationship graph information corresponding to the template failure label features;
the STEP1603 is used for outputting a fault repairing scheme based on a fault repairing scheme output unit of a fault repairing scheme recommendation model by combining the template fault causal graph information to generate fault repairing scheme output information corresponding to each fault label node in the template fault label characteristics;
STEP1604, combining the output information of the failure recovery scheme corresponding to each failure tag node in the template failure tag characteristics and the information of the template failure recovery scheme corresponding to each template failure tag node in the template failure tag characteristics, and updating the weight parameter layer of the failure recovery scheme recommendation model.
One implementation of STEP1601 may be: acquiring the template fault repairing training data cluster; the template fault repairing training data cluster covers template fault label characteristics and component fault symptom data corresponding to the template fault label characteristics; combining the template fault label characteristics and component fault symptom data corresponding to the template fault label characteristics to generate a template fault diagnosis map corresponding to the template fault label characteristics; the template fault diagnosis map represents a fault diagnosis entity corresponding to a fault diagnosis parameter in the template fault label characteristic; combining the template fault diagnosis map, predicting the fault diagnosis probability based on a fault diagnosis probability prediction unit, and generating a fault diagnosis probability value of a template fault diagnosis parameter corresponding to the template fault label characteristic; combining the fault diagnosis probability value of the template fault diagnosis parameters, respectively extracting fault diagnosis parameters from each fault label node in the template fault label characteristics, and generating a fault diagnosis parameter sequence of each fault label node of the template fault label characteristics; generating a fault repairing instance map by combining the fault repairing instances of the fault diagnosis parameter sequence of each fault label node of each template fault label characteristic in the template fault repairing training data cluster; the fault repairing instance map represents a fault repairing instance linkage relation of the fault repairing mode; generating a fault repair mode associated with the fault repair scheme output unit by combining the linkage relation of the fault repair examples of the fault repair mode; the failure repair mode covers a plurality of failure repair refinement modes; and combining the fault diagnosis parameter sequence of each fault label node in the template fault label characteristics, recommending a fault repair scheme based on the fault repair mode, and generating template fault repair scheme information associated with each fault label node in the template fault label characteristics.
According to the same inventive concept, an embodiment of the present invention further provides a fault detection server, and referring to fig. 2, fig. 2 is a structural diagram of the fault detection server 100 provided in the embodiment of the present invention, and the fault detection server 100 may generate a relatively large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 112 (e.g., one or more processors) and a memory 111. Wherein the memory 111 may be a transient storage or a persistent storage. The program stored in the memory 111 may include one or more modules, each of which may include a series of instructions operating on the fault detection server 100. Further, the central processor 112 may be configured to communicate with the memory 111, and execute a series of instruction operations in the memory 111 on the failure detection server 100.
The failure detection server 100 may also include one or more power supplies, one or more communication units 113, one or more pass-to-output interfaces, and/or one or more operating systems, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and the like.
The steps performed by the failure detection server in the above embodiments may be according to the failure detection server structure shown in fig. 2.
In addition, a machine-readable medium for storing a computer program for executing the method provided by the above embodiment is also provided.
Embodiments of the present application further provide a computer program product including instructions, which when executed on a computer, cause the computer to execute the method provided by the above embodiments.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program code, when executed by the processor or controller, causes the functions/acts specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
As used in this disclosure, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user may provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
Claims (10)
1. A solar cell fault analysis method is executed based on a fault detection server, and is characterized by comprising the following steps:
acquiring state flow direction characteristic distribution corresponding to target battery operation abnormal data uploaded by a solar battery monitoring system; the target battery abnormal operation data is data with abnormal operation activities obtained by monitoring state data of a target solar battery system; the target solar cell system is a cell target to be subjected to fault analysis;
performing member activity positioning on the state flow direction characteristic distribution, and outputting a first member state flow direction characteristic distribution corresponding to each battery operation abnormal activity in the target battery operation abnormal data;
loading the first member state flow direction characteristic distribution into a first target member fault analysis model associated with each first member state flow direction characteristic distribution, and outputting a first fault classification confidence corresponding to the target battery operation abnormal data; the first fault classification confidence degree comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic;
loading the state flow direction characteristic distribution to a target overall fault analysis model, and outputting a second fault classification confidence corresponding to the target battery operation abnormal data; the second fault classification confidence degree comprises the support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic;
and determining corresponding fault label characteristics of the target solar cell system based on the first fault classification confidence and the second fault classification confidence.
2. The method for analyzing the faults of the solar cell according to claim 1, wherein the loading each of the first member state flow direction feature distributions into a first target member fault analysis model associated with each of the first member state flow direction feature distributions and outputting a first fault classification confidence degree corresponding to the target cell operation abnormality data includes:
loading each first member state flow direction characteristic distribution into the first target member fault analysis model associated with each first member state flow direction characteristic distribution, and outputting a fault classification confidence coefficient corresponding to each first member state flow direction characteristic distribution;
fusing the flow direction characteristic distribution of each first member state according to the time-space domain information of the flow direction characteristic distribution of each first member state, and outputting the fused flow direction characteristic distribution of the first member state;
generating a fault classification confidence corresponding to the state flow direction feature distribution based on the fault classification confidence corresponding to each fused first member state flow direction feature distribution; the fault classification confidence corresponding to the state flow direction feature distribution comprises the support degree of each battery operation abnormal activity in the state flow direction feature distribution to each fault label feature;
and generating the first fault classification confidence coefficient based on the fault classification confidence coefficient corresponding to the state flow direction feature distribution.
3. The method of claim 1, wherein the determining the corresponding fault label feature of the target solar cell system based on the first fault classification confidence and the second fault classification confidence comprises:
adding the first fault classification confidence coefficient and the second fault classification confidence coefficient, and outputting a target fault classification confidence coefficient corresponding to the target battery operation abnormal data; the target fault classification confidence degree comprises a target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic;
generating a fault label characteristic corresponding to the target solar cell system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data aiming at each fault label characteristic;
generating a fault label feature corresponding to the target solar cell system based on the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label feature, including:
generating a maximum target support degree corresponding to each battery operation abnormal activity in the target battery operation abnormal data in the target support degree of each battery operation abnormal activity in the target battery operation abnormal data for each fault label characteristic;
outputting the fault label characteristics corresponding to the maximum target support degrees as the fault label characteristics corresponding to the abnormal operation activities of the batteries in the target battery operation abnormal data;
and generating a fault label characteristic corresponding to the target solar cell system based on the fault label characteristic corresponding to each battery operation abnormal activity in the target battery operation abnormal data.
4. The solar cell fault analysis method of claim 3, further comprising:
performing member activity positioning on the state flow direction characteristic distribution for multiple times under different characteristic dimensions, and outputting second member state flow direction characteristic distribution corresponding to the target battery operation abnormal data under at least one characteristic dimension; the distribution range of the second member state flow direction characteristic distribution is larger than the distribution range of the first member state flow direction characteristic distribution and smaller than the distribution range of the state flow direction characteristic distribution;
loading the second member state flow direction characteristics into respective associated second target member fault analysis models in a distributed manner, and outputting at least one third fault classification confidence degree corresponding to the target battery operation abnormal data; the at least one third fault classification confidence degree comprises the support degree of each fault label feature under the corresponding feature dimension of each battery operation abnormal activity in the target battery operation abnormal data;
adding the first fault classification confidence, the second fault classification confidence and the at least one third fault classification confidence and outputting the target fault classification confidence.
5. The solar cell fault analysis method of claim 4, further comprising:
acquiring a member fault analysis model waiting for model weight optimization; the member fault analysis model waiting for model weight optimization comprises unit fault analysis models corresponding to different characteristic dimensions;
performing model weight information optimization on the member fault analysis model waiting for model weight optimization based on a gradient descent algorithm, and outputting a first member fault analysis model meeting the model convergence requirement; and/or performing model weight information optimization on the member fault analysis model waiting for model weight optimization based on a specified loss function, and outputting a second member fault analysis model meeting the model convergence requirement;
outputting a target member fault analysis model based on the first member fault analysis model and/or the second member fault analysis model; the target member fault analysis model comprises the first target member fault analysis model, the target overall fault analysis model and the second target member fault analysis model.
6. The solar cell fault analysis method according to claim 1, wherein the target overall fault analysis model comprises a fault comparison unit and a fault classification unit;
the step of loading the state flow direction characteristic distribution to a target overall fault analysis model and outputting a second fault classification confidence degree corresponding to the target battery operation abnormal data comprises the following steps:
loading the state flow direction characteristics to the fault comparison unit in a distributed manner, and outputting a target correlation parameter value between a fault correlation characteristic corresponding to each battery operation abnormal activity in the target battery operation abnormal data and a fault correlation characteristic corresponding to each fault label characteristic; the dimension of the fault association feature corresponding to each battery operation abnormal activity in the target battery operation abnormal data is equal to the dimension of the fault association feature corresponding to each fault label feature;
loading the target associated parameter value to the fault classification unit, and outputting the second fault classification confidence; the second fault classification confidence is obtained by loading the target associated parameter value to a normalized multi-classification function through the fault classification unit.
7. The solar cell fault analysis method of claim 1, further comprising:
according to the corresponding fault label characteristics of the target solar cell system, fault repair is carried out on the target solar cell system, and the specific steps comprise:
inputting the fault label characteristics corresponding to the target solar cell system into a fault repair scheme recommendation model, and outputting fault repair scheme information corresponding to the target solar cell system;
and performing fault repair on the target solar cell system based on the fault repair scheme information.
8. The method of claim 7, wherein the step of training the fault remediation scheme recommendation model comprises:
extracting template fault label characteristics from a fault repair template database, and template fault repair scheme information corresponding to each corresponding fault label node in the template fault label characteristics;
extracting the failure causal relationship graph of the template failure label characteristics by combining a failure causal relationship graph extraction unit of a failure repair scheme recommendation model, and outputting template failure causal relationship graph information corresponding to the template failure label characteristics;
combining the template fault causal relationship diagram information, outputting a fault repair scheme based on a fault repair scheme output unit of a fault repair scheme recommendation model, and generating fault repair scheme output information corresponding to each fault label node in the template fault label characteristics;
updating a weight parameter layer of the failure recovery scheme recommendation model by combining the failure recovery scheme output information corresponding to each failure tag node in the template failure tag characteristics and the template failure recovery scheme information corresponding to each template failure tag node in the template failure tag characteristics;
the extracting of the template fault label characteristics from the fault repairing template database and the template fault repairing scheme information corresponding to each corresponding fault label node in the template fault label characteristics comprise:
acquiring a template fault repairing training data cluster; the template fault repairing training data cluster covers template fault label characteristics and component fault symptom data corresponding to the template fault label characteristics;
combining the template fault label characteristics and component fault symptom data corresponding to the template fault label characteristics to generate a template fault diagnosis map corresponding to the template fault label characteristics; the template fault diagnosis map represents a fault diagnosis entity corresponding to a fault diagnosis parameter in the template fault label characteristic;
combining the template fault diagnosis map, predicting the fault diagnosis probability based on a fault diagnosis probability prediction unit, and generating a fault diagnosis probability value of a template fault diagnosis parameter corresponding to the template fault label characteristic;
extracting fault diagnosis parameters from each fault label node in the template fault label characteristics respectively by combining the fault diagnosis probability value of the template fault diagnosis parameters, and generating a fault diagnosis parameter sequence of each fault label node of the template fault label characteristics;
generating a fault repairing instance map by combining the fault repairing instances of the fault diagnosis parameter sequence of each fault label node of each template fault label characteristic in the template fault repairing training data cluster; the fault repairing instance map represents a fault repairing instance linkage relation of a fault repairing mode;
generating a fault repairing mode associated with the fault repairing scheme output unit by combining the linkage relation of the fault repairing examples of the fault repairing mode; the failure repair mode covers a plurality of failure repair refinement modes;
and combining the fault diagnosis parameter sequence of each fault label node in the template fault label characteristics, recommending a fault repair scheme based on the fault repair mode, and generating template fault repair scheme information associated with each fault label node in the template fault label characteristics.
9. A fault detection server, comprising:
a processor;
a memory having stored therein a computer program that, when executed, implements the solar cell failure analysis method of any of claims 1-8.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed, implements the solar cell failure analysis method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210700902.0A CN114785284B (en) | 2022-06-21 | 2022-06-21 | Solar cell fault analysis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210700902.0A CN114785284B (en) | 2022-06-21 | 2022-06-21 | Solar cell fault analysis method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114785284A true CN114785284A (en) | 2022-07-22 |
| CN114785284B CN114785284B (en) | 2022-08-23 |
Family
ID=82422029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210700902.0A Active CN114785284B (en) | 2022-06-21 | 2022-06-21 | Solar cell fault analysis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114785284B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115081552A (en) * | 2022-07-28 | 2022-09-20 | 一道新能源科技(衢州)有限公司 | Solar cell data exception handling method and system based on cloud platform |
| CN116430817A (en) * | 2023-04-26 | 2023-07-14 | 同心县启胜新能源科技有限公司 | Data acquisition processing method and system applied to photovoltaic module production system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010245431A (en) * | 2009-04-09 | 2010-10-28 | Lasertec Corp | Evaluating apparatus and evaluating method for solar battery and manufacturing method for solar battery |
| CN108197774A (en) * | 2017-12-08 | 2018-06-22 | 囯网河北省电力有限公司电力科学研究院 | A kind of abnormality diagnostic method and device of distributed photovoltaic power generation amount |
| CN112163008A (en) * | 2020-09-28 | 2021-01-01 | 邓燕平 | Big data analysis-based user behavior data processing method and cloud computing platform |
| CN113792088A (en) * | 2021-09-10 | 2021-12-14 | 阳光电源股份有限公司 | Fault detection method and device for photovoltaic string and storage medium |
| CN114169523A (en) * | 2022-02-10 | 2022-03-11 | 一道新能源科技(衢州)有限公司 | Solar cell use data analysis method and system |
-
2022
- 2022-06-21 CN CN202210700902.0A patent/CN114785284B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010245431A (en) * | 2009-04-09 | 2010-10-28 | Lasertec Corp | Evaluating apparatus and evaluating method for solar battery and manufacturing method for solar battery |
| CN108197774A (en) * | 2017-12-08 | 2018-06-22 | 囯网河北省电力有限公司电力科学研究院 | A kind of abnormality diagnostic method and device of distributed photovoltaic power generation amount |
| CN112163008A (en) * | 2020-09-28 | 2021-01-01 | 邓燕平 | Big data analysis-based user behavior data processing method and cloud computing platform |
| CN113792088A (en) * | 2021-09-10 | 2021-12-14 | 阳光电源股份有限公司 | Fault detection method and device for photovoltaic string and storage medium |
| CN114169523A (en) * | 2022-02-10 | 2022-03-11 | 一道新能源科技(衢州)有限公司 | Solar cell use data analysis method and system |
Non-Patent Citations (1)
| Title |
|---|
| 李振兴等: "含分布式电源与随机负荷的主动配电网保护", 《高电压技术》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115081552A (en) * | 2022-07-28 | 2022-09-20 | 一道新能源科技(衢州)有限公司 | Solar cell data exception handling method and system based on cloud platform |
| CN116430817A (en) * | 2023-04-26 | 2023-07-14 | 同心县启胜新能源科技有限公司 | Data acquisition processing method and system applied to photovoltaic module production system |
| CN116430817B (en) * | 2023-04-26 | 2023-09-29 | 同心县启胜新能源科技有限公司 | Data acquisition processing method and system applied to photovoltaic module production system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114785284B (en) | 2022-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114785284B (en) | Solar cell fault analysis method and system | |
| Xie et al. | A review of machine learning applications in power system resilience | |
| US20200210854A1 (en) | System and method for fault detection of components using information fusion technique | |
| CN110297911A (en) | Internet of Things (IOT) calculates the method and system that cognition data are managed and protected in environment | |
| Mohammadi et al. | Machine learning assisted stochastic unit commitment during hurricanes with predictable line outages | |
| Contreras-Jiménez et al. | Multi-state system reliability analysis of HVDC transmission systems using matrix-based system reliability method | |
| CN114580263A (en) | Knowledge graph-based information system fault prediction method and related equipment | |
| CN116155956B (en) | Multiplexing communication method and system based on gradient decision tree model | |
| CN116169675B (en) | Power system dynamic stability online evaluation method considering operation mode change | |
| CN111027591B (en) | Node fault prediction method for large-scale cluster system | |
| CN116245033B (en) | Artificial intelligent driven power system analysis method and intelligent software platform | |
| CN116502771B (en) | Power distribution method and system based on electric power material prediction | |
| Kayode et al. | Lirul: A lightweight lstm based model for remaining useful life estimation at the edge | |
| CN116755974A (en) | Cloud computing platform operation and maintenance method and device, electronic equipment and storage medium | |
| CN116225848A (en) | Log monitoring method, device, equipment and medium | |
| CN115563507A (en) | Generation method, device and equipment for renewable energy power generation scene | |
| Mayadevi et al. | SCADA-based operator support system for power plant equipment fault forecasting | |
| Aragon et al. | Decision Support System for Photovoltaic Fault Detection Avoiding Meteorological Conditions | |
| Ahmadi et al. | The islanded micro‐grid large signal stability analysis based on neuro‐fuzzy model | |
| Ku et al. | Self-learning mechanism for prediction of energy consumption and generation | |
| Li et al. | Multi-source heterogeneous log fusion technology of power information system based on big data and imprecise reasoning theory | |
| Li et al. | Power system transient voltage vulnerability assessment based on knowledge visualization of CNN | |
| Qu et al. | A Reliability Assessment Method Based on Convolutional Neural Network | |
| CN117349747A (en) | Offline fault reason classification method for electric power Internet of things intelligent terminal | |
| CN118052650A (en) | Shared energy storage transaction method, system, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221208 Address after: No. 108, Yaojia Road, Hailing District, Taizhou City, Jiangsu Province, 225300 Patentee after: Yidao new energy technology (Taizhou) Co.,Ltd. Address before: 324022 room 604, building 3, Donggang Third Road, green industry cluster, Qujiang District, Quzhou City, Zhejiang Province Patentee before: A New Energy Technology (Quzhou) Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230228 Address after: No. 43, bailing South Road, Donggang District, green industry cluster district, Quzhou City, Zhejiang Province Patentee after: A New Energy Technology (Quzhou) Co.,Ltd. Address before: No. 108, Yaojia Road, Hailing District, Taizhou City, Jiangsu Province, 225300 Patentee before: Yidao new energy technology (Taizhou) Co.,Ltd. |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: No. 43 Bailing South Road, Quzhou City, Zhejiang Province 324022 Patentee after: Yidao New Energy Technology Co.,Ltd. Address before: No. 43, bailing South Road, Donggang District, green industry cluster district, Quzhou City, Zhejiang Province Patentee before: A New Energy Technology (Quzhou) Co.,Ltd. |