CN115940809A - Solar panel fault detection method based on power data and visual analysis - Google Patents
Solar panel fault detection method based on power data and visual analysis Download PDFInfo
- Publication number
- CN115940809A CN115940809A CN202310222179.4A CN202310222179A CN115940809A CN 115940809 A CN115940809 A CN 115940809A CN 202310222179 A CN202310222179 A CN 202310222179A CN 115940809 A CN115940809 A CN 115940809A
- Authority
- CN
- China
- Prior art keywords
- solar panel
- power
- real
- time
- imaging
- 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
- 230000000007 visual effect Effects 0.000 title claims abstract description 87
- 238000004458 analytical method Methods 0.000 title claims abstract description 50
- 238000001514 detection method Methods 0.000 title claims abstract description 47
- 238000003384 imaging method Methods 0.000 claims abstract description 130
- 230000002159 abnormal effect Effects 0.000 claims abstract description 92
- 238000003745 diagnosis Methods 0.000 claims abstract description 63
- 230000004927 fusion Effects 0.000 claims abstract description 46
- 238000000034 method Methods 0.000 claims description 61
- 238000000605 extraction Methods 0.000 claims description 60
- 230000006870 function Effects 0.000 claims description 32
- 238000012952 Resampling Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 238000011176 pooling Methods 0.000 claims description 9
- 230000007547 defect Effects 0.000 claims description 8
- 238000005215 recombination Methods 0.000 claims description 7
- 230000006798 recombination Effects 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000005401 electroluminescence Methods 0.000 claims description 3
- 238000003331 infrared imaging Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 7
- 150000001875 compounds Chemical class 0.000 abstract description 5
- 230000005611 electricity Effects 0.000 abstract description 5
- 238000010248 power generation Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 12
- 230000005856 abnormality Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 9
- 238000012423 maintenance Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000002372 labelling Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000000306 component Substances 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000011049 filling Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 206010027146 Melanoderma Diseases 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008521 reorganization Effects 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- 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
Landscapes
- Photovoltaic Devices (AREA)
Abstract
The invention discloses a solar panel fault detection method based on power data and visual analysis, which belongs to the technical field of solar cells and solves the problem that the fault judgment accuracy is insufficient because the conventional solar panel fault detection method cannot be used for accurately analyzing and judging the solar panel fault by combining the solar panel running power and visual imaging, and comprises the following steps: acquiring real-time operation power data of the solar panel; obtaining a power abnormal characteristic set; acquiring a solar panel real-time visual imaging set, and performing fusion analysis diagnosis based on a fault fusion diagnosis model to obtain a solar panel fault diagnosis result; this application monitors the early warning through compound trouble fusion diagnostic model to solar panel, can in time discover solar panel's generated power unusual, can also combine solar panel generated power and real-time vision imaging set, carries out positioning analysis to solar panel's the unusual point of electricity generation, has made things convenient for the monitoring of solar panel trouble.
Description
Technical Field
The invention belongs to the technical field of solar cells, and particularly relates to a solar panel fault detection method based on power data and visual analysis.
Background
With the increasing shortage of energy and the attention on environmental protection problems, the conversion of light energy and solar energy by adopting the solar panel for power generation gradually becomes the main trend of new energy development, the solar panel is the core component of photovoltaic power generation, the solar panel can directly convert sunlight radiation into electric energy, and the stability and the power generation efficiency of a photovoltaic power generation system are influenced by the normal work of the solar panel.
The existing solar panel fault detection method mainly diagnoses the photovoltaic module through IV curve (current-voltage curve) data, but the existing solar panel fault detection method cannot combine solar panel running power and visual imaging to accurately analyze and judge the solar panel fault, so that the fault judgment accuracy is insufficient, and the power generation efficiency of a photovoltaic power generation system is influenced.
Disclosure of Invention
The invention aims to provide a solar panel fault detection method based on power data and visual analysis aiming at the defects of the prior art, and solves the problems that the existing solar panel fault detection method cannot be combined with solar panel running power and visual imaging to carry out accurate analysis and judgment on solar panel faults, so that the fault judgment accuracy is insufficient, and the power generation efficiency of a photovoltaic power generation system is influenced.
The existing solar panel fault detection method mainly diagnoses a photovoltaic module through IV curve (current-voltage curve) data, but the existing solar panel fault detection method cannot combine solar panel operating power and visual imaging to carry out accurate analysis and judgment on solar panel faults, so that the fault judgment accuracy is insufficient, and the power generation efficiency of a photovoltaic power generation system is affected. Acquiring real-time operation power data of the solar panel to obtain a power abnormal characteristic set; acquiring a real-time visual imaging set of the solar panel, constructing a fault fusion diagnosis model based on a Segnet network structure, and performing fusion analysis diagnosis on the power abnormal feature set and a pre-detected visual imaging set based on the fault fusion diagnosis model to obtain a fault diagnosis result of the solar panel; this application monitors the early warning to solar panel through compound trouble fusion diagnostic model to solar panel real-time operating power data and vision imaging set, can in time discover solar panel's generated power unusual, can also combine solar panel generated power and real-time vision imaging set, carries out positioning analysis to solar panel's the unusual point of electricity generation, has made things convenient for the monitoring of solar panel trouble, has improved solar panel's maintenance efficiency.
The invention is realized in this way, the solar panel fault detection method based on power data and visual analysis includes:
acquiring real-time operation power data of the solar panel, and preprocessing the real-time operation power data, wherein the real-time operation power data has time sequence, and comprises the temperature of the solar panel, the operation power of the solar panel, real-time voltage and real-time current;
extracting and screening the preprocessed real-time running power data based on a pre-trained abnormal power extraction model to obtain a power abnormal characteristic set;
the method comprises the steps of obtaining a solar panel real-time vision imaging set based on machine vision pre-detection, wherein the machine vision pre-detection is carried out by comparing the real-time vision imaging set with a standard vision imaging set, so that the pre-detection of the vision imaging set is realized;
and constructing a fault fusion diagnosis model based on the Segnet network structure, and performing fusion analysis diagnosis on the power abnormal feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model to obtain a solar panel fault diagnosis result.
Preferably, the method for extracting the abnormal power extraction model specifically includes:
acquiring multiple groups of normal-operation standard solar panel operation power data, and constructing an abnormal power extraction model based on the standard solar panel operation power data;
executing an abnormal power extraction model by taking the preprocessed real-time running power data as input, and normalizing the real-time running power data by the abnormal power extraction model;
and the abnormal power extraction model calls an abnormal power extraction function and extracts a power abnormal feature set in the real-time operation power data based on the abnormal power extraction function.
Preferably, the method for constructing the abnormal power extraction model based on the standard solar panel operating power data specifically includes:
acquiring N groups of normal operating standard solar panel operating power data;
and converting the N groups of samples into data matrixes at different moments t based on the temperature of the solar panel, the operating power of the solar panel, the real-time voltage and the real-time current category, and defining an abnormal power extraction function of converting the N groups of samples into the data matrixes at different moments t by using a multi-objective evolutionary algorithm.
Preferably, the method for extracting a set of power abnormality characteristics from real-time operating power data based on the abnormal power extraction function specifically includes:
an abnormal power extraction function is called, and a data matrix error vector, an average value error and a standard error of the standard solar panel operation power data are calculated;
and identifying a power abnormal feature set based on a least square method by taking the error vector, the average error and the standard error of the data matrix as input abnormal power extraction functions.
Preferably, the manner of acquiring the real-time visual imaging set of the solar panel is infrared imaging technology, electroluminescence imaging or visible light imaging.
Preferably, the method for acquiring a real-time visual imaging set of a solar panel specifically includes:
shooting based on a light imaging technology to obtain a plurality of groups of solar panel images, wherein the solar panel images comprise local images and global images;
and cutting, projecting and resampling the solar panel image to obtain a corrected real-time visual imaging set.
Preferably, the method for acquiring a real-time visual imaging set of a solar panel further comprises:
and grabbing the corrected real-time visual imaging set based on Python programming, and rasterizing the grabbed real-time visual imaging set to obtain a grid layer set with consistent size.
Preferably, the fault fusion diagnosis model based on the Segnet network structure includes:
a diagnostic decoder;
the diagnostic decoder comprises an imaging characteristic convolution layer and a characteristic pooling layer, wherein the imaging characteristic convolution layer is used for extracting image characteristics in an imaging set;
the characteristic pooling layer is used for converting and reducing the image characteristics in the imaging set;
and the recombination encoder is used for restoring and resampling image characteristics in the imaging set.
Preferably, the method for performing reduction resampling on image features in an imaging set by the restructuring encoder specifically includes:
acquiring the image features of the imaging set after the feature pooling layer processing, classifying the image features of the imaging set through Segnet to obtain a feature classification set, and calculating the feature coverage range through a formula for each feature Q (x, y) in the feature classification set:
wherein m is the number of grids of a single group of solar panels, D (x, y) is the brightness value of a characteristic Q (x, y), maxU is the threshold difference between a characteristic edge point and a central point, the value range is 0.2-2, \8706representsthe weight coefficient of the characteristic central point, and delta is the weight division rate of the characteristic;
the resulting calculation formula for the convolution of D (x, y) is:
fn (x, y) represents the feature center expression function, epsilon represents the luminance color factor, W is the feature weighting index, W ∈ (1, ∞).
Preferably, the method for performing fusion analysis diagnosis on the power abnormality feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model specifically includes:
performing recombination imaging on the visual imaging set based on the fault fusion diagnosis model to obtain a labeled imaging set labeled with defects;
a power abnormal feature set is called, and the power abnormal feature set is identified based on a least square method to obtain a power diagnosis result;
and matching the power diagnosis result with the labeled imaging set, and fusing, analyzing and diagnosing the fault type of the solar panel.
Compared with the prior art, the embodiment of the application mainly has the following beneficial effects:
this application monitors the early warning to solar panel through compound trouble fusion diagnostic model to solar panel real-time operating power data and vision imaging set, can in time discover solar panel's generated power unusual, can also combine solar panel generated power and real-time vision imaging set, carries out positioning analysis to solar panel's the unusual point of electricity generation, has made things convenient for the monitoring of solar panel trouble, has improved solar panel's maintenance efficiency.
Drawings
Fig. 1 is a schematic flow chart illustrating an implementation process of the solar panel fault detection method based on power data and visual analysis provided by the present invention.
FIG. 2 is a schematic flow chart of an implementation of the method for performing smoothness processing on a fitted curve of real-time operating power data according to the present invention.
Fig. 3 is a schematic implementation flow diagram of the abnormal power extraction model extraction method provided in the present invention.
Fig. 4 is a schematic flow chart of an implementation of the method for constructing the abnormal power extraction model based on the standard solar panel operating power data provided by the invention.
Fig. 5 is a schematic flow chart illustrating an implementation of the method for extracting a set of power anomaly characteristics from real-time operating power data based on an anomaly power extraction function according to the present invention.
Fig. 6 is a schematic flow chart illustrating the implementation of the method for acquiring a real-time visual imaging set of a solar panel according to the present invention.
Fig. 7 is a schematic flow chart of an implementation of the method for performing reduction resampling on image features in an imaging set by using a reorganization encoder provided by the present invention.
Fig. 8 is a schematic implementation flow diagram of a method for performing fusion analysis and diagnosis on a power abnormality feature set and a pre-detected visual imaging set based on a fault fusion diagnosis model according to the present invention.
Fig. 9 is a schematic structural diagram of a solar panel fault detection system based on power data and visual analysis according to the present invention.
FIG. 10 is a schematic diagram of a model diagnostic module according to the present invention.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "including" and "having," and any variations thereof, in the description and claims of this application and the description of the above figures are intended to cover non-exclusive inclusions. The terms "first," "second," and the like in the description and claims of this application or in the above-described drawings are used for distinguishing between different objects and not for describing a particular order.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The existing solar panel fault detection method mainly diagnoses a photovoltaic module through IV curve (current-voltage curve) data, but the existing solar panel fault detection method cannot combine solar panel operating power and visual imaging to carry out accurate analysis and judgment on solar panel faults, so that the fault judgment accuracy is insufficient, and the power generation efficiency of a photovoltaic power generation system is influenced. Acquiring real-time operation power data of the solar panel to obtain a power abnormal characteristic set; acquiring a solar panel real-time visual imaging set, constructing a fault fusion diagnosis model based on a Segnet network structure, and performing fusion analysis diagnosis on a power abnormal feature set and a pre-detected realization visual imaging set based on the fault fusion diagnosis model to obtain a solar panel fault diagnosis result; this application monitors the early warning to solar panel through compound trouble fusion diagnostic model to solar panel real-time running power data and vision formation of image collection, can in time discover that solar panel's generated power is unusual, can also combine solar panel generated power and real-time vision formation of image collection, carries out positioning analysis to solar panel's the unusual point of electricity generation, has made things convenient for the monitoring of solar panel trouble, has improved solar panel's maintenance efficiency.
The embodiment of the invention provides a solar panel fault detection method based on power data and visual analysis, and as shown in fig. 1, a schematic flow chart of the implementation of the solar panel fault detection method based on power data and visual analysis is shown, and the solar panel fault detection method based on power data and visual analysis specifically comprises the following steps:
step S10, acquiring real-time operation power data of the solar panel, and preprocessing the real-time operation power data, wherein the real-time operation power data have time sequence, and the real-time operation power data comprise the temperature of the solar panel, the operation power of the solar panel, real-time voltage and real-time current;
it should be noted that, in the present application, the real-time operation power data includes, but is not limited to, a temperature of the solar panel, an operation power of the solar panel, a real-time voltage, and a real-time current, and meanwhile, the collected real-time operation power data of the solar panel needs to be normalized by a linear function.
Step S20, extracting and screening the preprocessed real-time operation power data based on a pre-trained abnormal power extraction model to obtain a power abnormal feature set;
step S30, acquiring a real-time visual imaging set of the solar panel, wherein the real-time visual imaging set is based on machine visual pre-detection, and the machine visual pre-detection is carried out by comparing the real-time visual imaging set with a standard visual imaging set to realize the pre-detection of the visual imaging set;
and S40, constructing a fault fusion diagnosis model based on the Segnet network structure, and performing fusion analysis diagnosis on the power abnormal feature set and the pre-detected visual imaging realization set based on the fault fusion diagnosis model to obtain a solar panel fault diagnosis result.
In this embodiment, the real-time operating power data is preprocessed by means including, but not limited to, missing data processing and fitted curve smoothing processing of the real-time operating power data.
Illustratively, the missing data processing of the real-time operation power data is performed based on a preset gap filling rule, where the preset gap filling rule is as follows: and judging whether the missing degree of the power data of a certain dimension is greater than a preset missing threshold, if so, discarding the power data of the dimension, and if not, completing the missing part of the power data of the certain dimension.
In an optional embodiment of the present invention, as shown in fig. 2, when performing smoothness processing on the fit curve of the real-time operation power data, the method for performing smoothness processing on the fit curve of the real-time operation power data specifically includes:
step S101, introducing a processable rule smooth constraint frame, wherein a power data set of each dimension before smoothness processing is set as Z (x, y);
step S102, carrying out chromosome coding on each group of data of the power data set Z (x, y) to form a coding set Z a （x，y）；
Step S103, then, for the coding set Z a (x, y) setting a smooth fitting target of accuracy and reasonableness;
step S104, carrying out rule matching on the smooth fitting target, and then verifying the fitness of the corresponding rule;
step S105, if the fitness accords with the smooth fitting requirement, outputting a smooth optimal set Z corresponding to the coding set b (x, y). And if the fitness does not meet the smooth fitting requirement, repeating the steps.
This application monitors the early warning to solar panel through compound trouble fusion diagnostic model to solar panel real-time operating power data and vision imaging set, can in time discover solar panel's generated power unusual, can also combine solar panel generated power and real-time vision imaging set, carries out positioning analysis to solar panel's the unusual point of electricity generation, has made things convenient for the monitoring of solar panel trouble, has improved solar panel's maintenance efficiency.
The embodiment of the present invention provides an extraction method of an abnormal power extraction model, and as shown in fig. 3, an implementation flow diagram of the extraction method of the abnormal power extraction model is shown, where the extraction method of the abnormal power extraction model specifically includes:
step S201, acquiring multiple groups of normal operating standard solar panel operating power data, and constructing an abnormal power extraction model based on the standard solar panel operating power data;
step S202, taking preprocessed real-time running power data as input, executing an abnormal power extraction model, and normalizing the real-time running power data by the abnormal power extraction model;
step S203, the abnormal power extraction model calls an abnormal power extraction function, and a power abnormal feature set in the real-time operation power data is extracted based on the abnormal power extraction function.
In this embodiment, in order to ensure the robustness of the solar panel operating power data set, a long-short term memory network is adopted to quickly construct an abnormal power extraction model, meanwhile, the extraction of the power abnormal feature set in the real-time operating power data based on the abnormal power extraction function is to divide the real-time operating power data set in different scales at least twice, and the real-time operating power data set divided each time is subjected to the combination of the SSNBDL network structure and the least square method to identify the power abnormal feature set.
The embodiment of the invention provides a method for constructing an abnormal power extraction model based on standard solar panel operating power data, as shown in fig. 4, an implementation flow diagram of the method for constructing the abnormal power extraction model based on the standard solar panel operating power data is shown, and the method for constructing the abnormal power extraction model based on the standard solar panel operating power data specifically comprises the following steps:
step S2011, acquiring N groups of normal operating standard solar panel operating power data;
step S2012, converting the N groups of samples into data matrixes at different moments t based on the temperature of the solar panel, the operating power of the solar panel, the real-time voltage and the real-time current category, and defining an abnormal power extraction function of converting the N groups of samples into the data matrixes at different moments t through a multi-objective evolutionary algorithm.
It should be noted that the abnormal power extraction function is a gaussian filter function, the abnormal power extraction function can perform filtering processing on the data matrix, and the abnormal power extraction function is used to analyze and process the data matrix without destroying the structure of the matrix, thereby facilitating the retrieval of data and improving the anti-interference capability of the data matrix.
In the present embodiment, the expression formula of the abnormal power extraction function is:
wherein phi is a Gaussian matrix scale, the size of phi is K × L, and m and n are respectively the length and the width of the matrix.
The embodiment of the invention provides a method for extracting a power abnormal feature set in real-time running power data based on an abnormal power extraction function, and as shown in fig. 5, an implementation flow diagram of the method for extracting the power abnormal feature set in the real-time running power data based on the abnormal power extraction function is shown, and the method for extracting the power abnormal feature set in the real-time running power data based on the abnormal power extraction function specifically comprises the following steps:
step S2031, an abnormal power extraction function is called, and a data matrix error vector, an average error and a standard error of the standard solar panel operation power data are calculated;
step S2032, the error vector of the data matrix, the error of the average value and the standard error are used as input to an abnormal power extraction function, and a power abnormal feature set is identified based on a least square method.
It should be noted that, in the present application, the manner of acquiring the real-time visual imaging set of the solar panel is an infrared imaging technique, an electroluminescence imaging technique, or a visible light imaging technique. The visual imaging set comprises common internal and external defects, such as black core, black spot, short circuit black sheet, over-welding sheet, grid breaking sheet, light and shade sheet, hidden crack and other types of defects.
The embodiment of the invention provides a method for acquiring a real-time visual imaging set of a solar panel, and as shown in fig. 6, a schematic diagram of an implementation flow of the method for acquiring the real-time visual imaging set of the solar panel is shown, and the method for acquiring the real-time visual imaging set of the solar panel specifically comprises the following steps:
s301, shooting based on a light imaging technology, and acquiring a plurality of groups of solar panel images, wherein the solar panel images comprise local images and global images;
and S302, cutting, projecting and resampling the solar panel image to obtain a corrected real-time visual imaging set.
And S303, grabbing the corrected real-time visual imaging set based on Python programming, and rasterizing the grabbed real-time visual imaging set to obtain a grid layer set with consistent size.
In this embodiment the resolution of the set of uniformly sized raster layers is 30mm x 30mm.
In an optional embodiment of the present application, the fault fusion diagnosis model based on the Segnet network structure includes:
a diagnostic decoder;
the diagnostic decoder comprises an imaging characteristic convolution layer and a characteristic pooling layer, wherein the imaging characteristic convolution layer is used for extracting image characteristics in an imaging set;
the characteristic pooling layer is used for converting and reducing the image characteristics in the imaging set;
and the recombination encoder is used for restoring and resampling the image characteristics in the imaging set.
The embodiment of the invention provides a method for restoring and resampling image features in an imaging set by a recombined encoder, and as shown in fig. 7, an implementation flow schematic diagram of the method for restoring and resampling the image features in the imaging set by the recombined encoder is shown, and the method for restoring and resampling the image features in the imaging set by the recombined encoder specifically comprises the following steps:
step S4011, obtaining the image characteristics of the imaging set after the characteristic pooling layer processing;
step S4012, classifying the image features of the imaging set by Segnet to obtain a feature classification set, where each feature Q (x, y) in the feature classification set is calculated by using formula (2):
the method comprises the following steps of obtaining a characteristic Q (x, y), obtaining a value range of 87066, wherein m is the number of grids of a single-group solar panel, D (x, y) is the brightness value of the characteristic Q (x, y), maxU is the threshold difference between a characteristic edge point and a central point, the value range is 0.2-2, \\ 8706, representing the weight coefficient of the characteristic central point, and delta is the weight division rate of the characteristic;
the result of the convolution of D (x, y) is calculated as equation (3):
fn (x, y) represents the feature center expression function, epsilon represents the luminance color factor, W is the feature weighting index, W ∈ (1, ∞).
The embodiment of the present invention provides the method for performing fusion analysis diagnosis on the power abnormality feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model, and as shown in fig. 8, an implementation flow diagram of the method for performing fusion analysis diagnosis on the power abnormality feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model is shown, and the method for performing fusion analysis diagnosis on the power abnormality feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model specifically includes:
step S401, carrying out recombination imaging on the visual imaging set based on the fault fusion diagnosis model to obtain a labeled imaging set labeled with defects;
step S402, a power abnormal feature set is called, and the power abnormal feature set is identified based on a least square method to obtain a power diagnosis result;
and S403, matching the power diagnosis result with the labeled imaging set, and fusing, analyzing and diagnosing the fault type of the solar panel.
In this embodiment, the power anomaly feature set is called, the matching power diagnosis result and the labeled imaging set are parallel, the power anomaly feature point of the whole solar panel is traversed, and the labeled imaging set detection is performed on the currently processed feature point coordinates: and the solar panel meeting the power abnormality characteristic and marking the imaging set is in fault, and the position of the fault panel is fed back to a user.
The embodiment of the invention provides a solar panel fault detection system based on power data and visual analysis, as shown in fig. 9, which shows a schematic diagram of the solar panel fault detection system based on power data and visual analysis, and the solar panel fault detection system based on power data and visual analysis specifically comprises:
the power acquisition module 100, the power acquisition module 200 is configured to acquire real-time operating power data of the solar panel, and preprocess the real-time operating power data, where the real-time operating power data has a time sequence, and the real-time operating power data includes a temperature of the solar panel, an operating power of the solar panel, a real-time voltage, and a real-time current;
the abnormal feature set acquisition module 200 is used for extracting and screening the preprocessed real-time running power data based on a pre-trained abnormal power extraction model to obtain a power abnormal feature set;
the system comprises an imaging set acquisition module 300, wherein the imaging set acquisition module 300 acquires a real-time visual imaging set of the solar panel, and the real-time visual imaging set is based on machine vision pre-detection, wherein the machine vision pre-detection is carried out by comparing the real-time visual imaging set with a standard visual imaging set, so that the pre-detection of the visual imaging set is realized;
the model diagnosis module 400 is used for constructing a fault fusion diagnosis model based on a Segnet network structure, and performing fusion analysis diagnosis on the power abnormal feature set and the pre-detected realization visual imaging set based on the fault fusion diagnosis model to obtain a solar panel fault diagnosis result.
In this embodiment, the power obtaining module 100, the abnormal feature set obtaining module 200, the imaging set obtaining module 300, and the model diagnosing module 400 are connected in a 5G communication or DTU communication manner, and the power obtaining module 100 is connected to a plurality of groups of solar panels in a one-to-n manner, so as to implement real-time monitoring of the solar panels.
It should be noted that the power obtaining module 100 is connected to a data collecting terminal, the sensing range of the data collecting terminal is 10-300Hz, and the data collecting terminal can weaken various interference signals (such as high-frequency radiation, voltage fluctuation, etc.), adopt various anti-interference processing technologies, and can be used in the place with serious frequency conversion interference.
An embodiment of the present invention provides a model diagnosis module 400, as shown in fig. 10, which shows a schematic diagram of the model diagnosis module 400, where the model diagnosis module 400 specifically includes:
the labeling imaging unit 410, the labeling imaging unit 410 performs recombination imaging on the visual imaging set based on the fault fusion diagnosis model to obtain a labeling imaging set labeled with defects;
the power diagnosis unit 420 is used for calling a power abnormal feature set, identifying the power abnormal feature set based on a least square method and obtaining a power diagnosis result;
and the fault type judging unit 430, wherein the fault type judging unit 430 matches the power diagnosis result with the labeled imaging set, and performs fusion analysis and diagnosis on the fault type of the solar panel.
In this embodiment, the fault type determining unit 430 matches the power diagnosis result and the labeled imaging set, and integrates, analyzes and diagnoses the solar panel fault type to have surface defects of the solar panel, including hidden cracks, scratches, black spots, broken grids, holes and black corners, and meanwhile, the solar panel fault type can be used for dividing the power abnormality into solar panel positioning by combining with the power abnormality characteristic set, and the precise positioning of the single-group solar panel abnormality can be realized by matching with the labeled imaging set.
The embodiment of the present invention further provides a computer device, which includes a display screen, a memory, a processor and a computer program, where the memory stores the computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for detecting a solar panel fault based on power data and visual analysis.
It is understood that in the preferred embodiment of the present invention, the computer device may also be a notebook computer, a Personal Digital Assistant (PDA), a mobile phone, etc. capable of communicating.
An embodiment of the present invention further provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the processor is enabled to execute the steps of the method for detecting the failure of the solar panel based on the power data and the visual analysis.
Illustratively, a computer program can be partitioned into one or more modules, which are stored in memory and executed by a processor to implement the present invention. One or more of the modules may be a series of computer program instruction segments capable of performing certain functions, which are used to describe the execution of the computer program in the terminal device. For example, the computer program may be partitioned into units or modules of a solar panel failure detection system based on power data and visual analysis provided by the various system embodiments described above.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as software or hardware depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosed embodiments of the present invention.
Finally, it should be noted that the computer-readable storage medium (e.g., memory) herein can be either volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. By way of example, and not limitation, nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM), which can act as external cache memory. By way of example and not limitation, RAM may be available in a variety of forms such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The storage devices of the disclosed aspects are intended to comprise, without being limited to, these and other suitable types of memory.
In summary, the invention provides a solar panel fault detection method based on power data and visual analysis, the solar panel is monitored and early warned by a composite fault fusion diagnosis model for real-time operation power data and a visual imaging set of the solar panel, so that the abnormal power generation of the solar panel can be found in time, and the abnormal power generation point of the solar panel can be positioned and analyzed by combining the power generation of the solar panel and the real-time visual imaging set, thereby facilitating the monitoring of the solar panel fault and improving the maintenance efficiency of the solar panel.
It should be noted that for the sake of simplicity, the above-mentioned embodiments are all described as a series of acts or combinations, but those skilled in the art should understand that the present invention is not limited by the described acts or sequences, as some steps may be performed in other sequences or simultaneously according to the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no acts or modules are necessarily required in the present disclosure.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the above-described division of the units is only one type of division of logical functions, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or communication connection may be an indirect coupling or communication connection between devices or units through some interfaces, and may be in a telecommunication or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The above examples are only used to illustrate the technical solutions of the present invention, and do not limit the scope of the present invention. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from these embodiments without making any inventive step, fall within the scope of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art may still make various combinations, additions, deletions or other modifications of the features of the embodiments of the present invention according to the situation without conflict, so as to obtain different technical solutions without substantially departing from the spirit of the present invention, and these technical solutions also fall within the protection scope of the present invention.
Claims (10)
1. The solar panel fault detection method based on power data and visual analysis is characterized by comprising the following steps of:
acquiring real-time operation power data of the solar panel, and preprocessing the real-time operation power data, wherein the real-time operation power data has time sequence and comprises the temperature of the solar panel, the operation power of the solar panel, real-time voltage and real-time current;
extracting and screening the preprocessed real-time running power data based on a pre-trained abnormal power extraction model to obtain a power abnormal feature set;
the method comprises the steps of obtaining a solar panel real-time vision imaging set based on machine vision pre-detection, wherein the machine vision pre-detection is carried out by comparing the real-time vision imaging set with a standard vision imaging set, so that the pre-detection of the vision imaging set is realized;
and constructing a fault fusion diagnosis model based on the Segnet network structure, and performing fusion analysis diagnosis on the power abnormal feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model to obtain a solar panel fault diagnosis result.
2. The method of claim 1 for solar panel fault detection based on power data and visual analysis, wherein: the method for extracting the abnormal power extraction model specifically comprises the following steps:
acquiring multiple groups of normal operating standard solar panel operating power data, and constructing an abnormal power extraction model based on the standard solar panel operating power data;
executing an abnormal power extraction model by taking the preprocessed real-time running power data as input, and normalizing the real-time running power data by the abnormal power extraction model;
and the abnormal power extraction model calls an abnormal power extraction function and extracts a power abnormal feature set in the real-time operation power data based on the abnormal power extraction function.
3. The method of claim 2 for solar panel fault detection based on power data and visual analysis, wherein: the method for constructing the abnormal power extraction model based on the standard solar panel operating power data specifically comprises the following steps:
acquiring N groups of normal operating standard solar panel operating power data;
and converting the N groups of samples into data matrixes at different moments t based on the temperature of the solar panel, the operating power of the solar panel, the real-time voltage and the real-time current category, and defining an abnormal power extraction function of converting the N groups of samples into the data matrixes at different moments t by using a multi-objective evolutionary algorithm.
4. The method of claim 3 for solar panel fault detection based on power data and visual analysis, wherein: the method for extracting the power abnormal feature set in the real-time operation power data based on the abnormal power extraction function specifically comprises the following steps:
an abnormal power extraction function is called, and a data matrix error vector, an average error and a standard error of the standard solar panel operation power data are calculated;
and identifying a power abnormal feature set based on a least square method by taking the error vector, the average error and the standard error of the data matrix as input abnormal power extraction functions.
5. The method of any of claims 1-4 for solar panel fault detection based on power data and visual analysis, wherein: the mode for acquiring the real-time visual imaging set of the solar panel is an infrared imaging technology, electroluminescence imaging or visible light imaging.
6. The method of claim 4 for solar panel fault detection based on power data and visual analysis, wherein: the method for acquiring the real-time visual imaging set of the solar panel specifically comprises the following steps:
shooting based on a light imaging technology to obtain a plurality of groups of solar panel images, wherein the solar panel images comprise local images and global images;
and cutting, projecting and resampling the solar panel image to obtain a corrected real-time visual imaging set.
7. The method of claim 6 for solar panel fault detection based on power data and visual analysis, wherein: the method for acquiring the real-time visual imaging set of the solar panel further comprises the following steps:
and grabbing the corrected real-time visual imaging set based on Python programming, and rasterizing the grabbed real-time visual imaging set to obtain a grid layer set with consistent size.
8. The method of any of claims 1-4 for solar panel fault detection based on power data and visual analysis, wherein: the fault fusion diagnosis model based on the Segnet network structure comprises the following steps:
a diagnostic decoder;
the diagnostic decoder comprises an imaging characteristic convolution layer and a characteristic pooling layer, wherein the imaging characteristic convolution layer is used for extracting image characteristics in an imaging set;
the characteristic pooling layer is used for converting and reducing the image characteristics in the imaging set;
and the recombination encoder is used for restoring and resampling the image characteristics in the imaging set.
9. The method of claim 8 for solar panel fault detection based on power data and visual analysis, wherein: the method for performing reduction resampling on image features in an imaging set by the recombined encoder specifically comprises the following steps:
obtaining the image characteristics of the imaging set after the characteristic pooling layer processing, classifying the image characteristics of the imaging set through Segnet to obtain a characteristic classification set, wherein each characteristic in the characteristic classification set
Calculating the characteristic coverage range by a formula:
wherein ,
is the number of grids of a single group of solar panels>
Is characterized by>
Is based on the brightness value of->
Is the threshold difference between the characteristic edge point and the central point, and has the value range of 0.2-2->
Weight coefficient representing a characteristic center point, based on a characteristic number of pixels in a pixel>
A weight split ratio for the feature;
10. the method of claim 9 for solar panel fault detection based on power data and visual analysis, wherein: the method for performing fusion analysis diagnosis on the power abnormal feature set and the pre-detected visual imaging set based on the fault fusion diagnosis model specifically comprises the following steps:
performing recombination imaging on the visual imaging set based on the fault fusion diagnosis model to obtain a labeled imaging set labeled with defects;
calling a power abnormal feature set, and identifying the power abnormal feature set based on a least square method to obtain a power diagnosis result;
and matching the power diagnosis result with the labeled imaging set, and fusing, analyzing and diagnosing the fault type of the solar panel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310222179.4A CN115940809B (en) | 2023-03-09 | 2023-03-09 | Solar panel fault detection method based on power data and visual analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310222179.4A CN115940809B (en) | 2023-03-09 | 2023-03-09 | Solar panel fault detection method based on power data and visual analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115940809A true CN115940809A (en) | 2023-04-07 |
| CN115940809B CN115940809B (en) | 2023-06-09 |
Family
ID=86556127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310222179.4A Active CN115940809B (en) | 2023-03-09 | 2023-03-09 | Solar panel fault detection method based on power data and visual analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115940809B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116227743A (en) * | 2023-05-06 | 2023-06-06 | 中国华能集团清洁能源技术研究院有限公司 | Photovoltaic power generation power abnormal rate determining method and system based on tuna swarm algorithm |
| CN116955965A (en) * | 2023-09-20 | 2023-10-27 | 山东鑫泰莱光电股份有限公司 | Solar data fault prediction method, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647716A (en) * | 2018-05-09 | 2018-10-12 | 北京理工大学 | A kind of diagnosing failure of photovoltaic array method based on composite information |
| CN114862814A (en) * | 2022-05-18 | 2022-08-05 | 上海师范大学天华学院 | Solar cell panel defect detection method and system, storage medium and terminal |
| CN115225032A (en) * | 2022-07-19 | 2022-10-21 | 中国华电科工集团有限公司 | Distributed photovoltaic operation and maintenance system and operation and maintenance method |
| CN115712873A (en) * | 2022-11-18 | 2023-02-24 | 国网江苏省电力有限公司徐州供电分公司 | Photovoltaic grid-connected operation abnormity detection system and method based on data analysis and infrared image information fusion |
| CN115714580A (en) * | 2022-10-08 | 2023-02-24 | 北京鑫泰绿能科技有限公司 | Abnormity detection method for photovoltaic cell panel |
-
2023
- 2023-03-09 CN CN202310222179.4A patent/CN115940809B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647716A (en) * | 2018-05-09 | 2018-10-12 | 北京理工大学 | A kind of diagnosing failure of photovoltaic array method based on composite information |
| CN114862814A (en) * | 2022-05-18 | 2022-08-05 | 上海师范大学天华学院 | Solar cell panel defect detection method and system, storage medium and terminal |
| CN115225032A (en) * | 2022-07-19 | 2022-10-21 | 中国华电科工集团有限公司 | Distributed photovoltaic operation and maintenance system and operation and maintenance method |
| CN115714580A (en) * | 2022-10-08 | 2023-02-24 | 北京鑫泰绿能科技有限公司 | Abnormity detection method for photovoltaic cell panel |
| CN115712873A (en) * | 2022-11-18 | 2023-02-24 | 国网江苏省电力有限公司徐州供电分公司 | Photovoltaic grid-connected operation abnormity detection system and method based on data analysis and infrared image information fusion |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116227743A (en) * | 2023-05-06 | 2023-06-06 | 中国华能集团清洁能源技术研究院有限公司 | Photovoltaic power generation power abnormal rate determining method and system based on tuna swarm algorithm |
| CN116227743B (en) * | 2023-05-06 | 2023-09-01 | 中国华能集团清洁能源技术研究院有限公司 | Photovoltaic power generation power abnormal rate determining method and system based on tuna swarm algorithm |
| CN116955965A (en) * | 2023-09-20 | 2023-10-27 | 山东鑫泰莱光电股份有限公司 | Solar data fault prediction method, equipment and storage medium |
| CN116955965B (en) * | 2023-09-20 | 2024-02-02 | 山东鑫泰莱光电股份有限公司 | Solar data fault prediction method, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115940809B (en) | 2023-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115940809B (en) | Solar panel fault detection method based on power data and visual analysis | |
| CN108647716B (en) | Photovoltaic array fault diagnosis method based on composite information | |
| CN103886294B (en) | Method and system for recognizing number of movement times of arrester for convertor station based on images | |
| CN112734692A (en) | Transformer equipment defect identification method and device | |
| CN110619479B (en) | Fault judgment model modeling method and fault judgment method for photovoltaic module | |
| CN114966467A (en) | Power transmission line state evaluation method based on digital twinning | |
| JP2019022251A (en) | Solar cell diagnosis method and solar cell diagnosis system | |
| CN110362907A (en) | Based on the identification of SSD neural network transmission line of electricity target defect and diagnostic method | |
| CN116054417B (en) | Monitoring system and method for transformer substation | |
| CN115908407B (en) | Power equipment defect detection method and device based on infrared image temperature value | |
| CN115601197A (en) | Abnormal state detection method and device for photovoltaic power station | |
| CN115102287A (en) | Intelligent management and control system for new energy power station centralized area | |
| CN113191977A (en) | Image enhancement system for target detection and identification under severe environment condition | |
| CN117372954A (en) | Charging station safety monitoring method, device, system and computer equipment | |
| CN116488578A (en) | Photovoltaic equipment analysis method based on AI visual assistance | |
| CN116388685A (en) | Solar cell panel operation control system | |
| CN115954291A (en) | Crack monitoring system of TOPCon structure photovoltaic cell panel | |
| CN110045691A (en) | A kind of multitasking fault monitoring method of multi-source heterogeneous big data | |
| CN115311539A (en) | Overhead transmission line defect identification method, device, equipment and storage medium | |
| CN113850773A (en) | Detection method, device, equipment and computer readable storage medium | |
| CN218734201U (en) | Photovoltaic module fault diagnosis equipment | |
| CN114048815B (en) | Power grid operation information sensing system and sensing method based on plant side | |
| Gu et al. | Object detection of overhead transmission lines based on improved YOLOv5s | |
| CN118154609A (en) | Photovoltaic module fault detection method | |
| CN116760360A (en) | Energy loss monitoring system of photovoltaic power generation system |
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 |