CN112455676A - Intelligent monitoring and analyzing system and method for health state of photovoltaic panel - Google Patents
Intelligent monitoring and analyzing system and method for health state of photovoltaic panel Download PDFInfo
- Publication number
- CN112455676A CN112455676A CN201910846118.9A CN201910846118A CN112455676A CN 112455676 A CN112455676 A CN 112455676A CN 201910846118 A CN201910846118 A CN 201910846118A CN 112455676 A CN112455676 A CN 112455676A
- Authority
- CN
- China
- Prior art keywords
- photovoltaic panel
- data
- photovoltaic
- image
- panel array
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000036541 health Effects 0.000 title claims abstract description 27
- 238000011156 evaluation Methods 0.000 claims abstract description 15
- 238000013480 data collection Methods 0.000 claims abstract description 10
- 238000004891 communication Methods 0.000 claims abstract description 8
- 230000002159 abnormal effect Effects 0.000 claims description 24
- 230000014509 gene expression Effects 0.000 claims description 22
- 238000004140 cleaning Methods 0.000 claims description 12
- 239000000523 sample Substances 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000002372 labelling Methods 0.000 claims description 8
- 238000011176 pooling Methods 0.000 claims description 8
- 230000007613 environmental effect Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 6
- 230000003862 health status Effects 0.000 claims description 4
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 210000001503 joint Anatomy 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 description 27
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 208000003464 asthenopia Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003889 chemical engineering Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005401 electroluminescence Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64C—AEROPLANES; HELICOPTERS
- B64C39/00—Aircraft not otherwise provided for
- B64C39/02—Aircraft not otherwise provided for characterised by special use
- B64C39/024—Aircraft not otherwise provided for characterised by special use of the remote controlled vehicle type, i.e. RPV
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
-
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64U—UNMANNED AERIAL VEHICLES [UAV]; EQUIPMENT THEREFOR
- B64U2101/00—UAVs specially adapted for particular uses or applications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64U—UNMANNED AERIAL VEHICLES [UAV]; EQUIPMENT THEREFOR
- B64U2101/00—UAVs specially adapted for particular uses or applications
- B64U2101/30—UAVs specially adapted for particular uses or applications for imaging, photography or videography
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Aviation & Aerospace Engineering (AREA)
- Image Analysis (AREA)
- Photovoltaic Devices (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a photovoltaic panel health state intelligent monitoring and analyzing system and a method, wherein the system comprises: the system comprises a data collector and a monitoring server which are in communication connection, wherein the data collector is used for obtaining photovoltaic panel array data, and the monitoring server is used for deploying a Fast R-CNN algorithm to analyze the photovoltaic panel array data, evaluating the health state of a photovoltaic panel and determining an evaluation result; wherein, data collection station includes: the camera is arranged on the unmanned aerial vehicle and used for acquiring image data of the photovoltaic panel array, and the method comprises the following steps: acquiring photovoltaic panel array data by using a data acquisition unit; deploying a Fast R-CNN algorithm through a monitoring server to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and obtaining an evaluation result; wherein, data collection station includes: set up the camera on unmanned aerial vehicle, photovoltaic board array data include: through the image data that set up the camera on unmanned aerial vehicle and acquire, solved and patrolled and examined the not enough, the real-time poor, the inefficiency problem of precision in present photovoltaic power plant.
Description
Technical Field
The invention relates to the field of fault diagnosis of a photovoltaic panel system, in particular to a system and a method for intelligently detecting and analyzing the health state of a photovoltaic panel.
Background
With the rapid development of the photovoltaic industry, the number and scale of photovoltaic power stations are continuously increased, and the traditional inspection mode combining fixed-point monitoring and manual inspection cannot be competent for large-scale inspection of photovoltaic power stations. Because the monitoring camera has a monitoring dead angle and the imaging effect of a remote target is not good, the subjective interference of manual inspection is large, the visual fatigue is easy to generate, the accurate positioning and real-time transmission of a photovoltaic power station cannot be met, and the damage conditions of internal structures such as hidden cracks of a photovoltaic panel cannot be known. Unmanned aerial vehicle inspection is used as a safe and efficient inspection technology and is widely applied to a plurality of fields such as forest fire inspection, power line inspection, oil field inspection, meteorological investigation and the like in recent years; ultrasonic detection is used as a detection technology for nondestructively probing internal and external defects of equipment and is widely applied to the fields of metallurgy manufacturing, processing chemical engineering, aerospace, railway traffic and the like.
The inspection technology applied by the existing photovoltaic industry is not mature enough, the consideration during the use and design of the inspection technology is not thorough enough, the problems of insufficient inspection precision, poor real-time performance and low efficiency in a photovoltaic power station often exist, the inspection technology is complicated and troublesome to use, the existing researchers do not develop a good inspection technology, and the problems of insufficient inspection precision, poor real-time performance, low efficiency and the like in the existing photovoltaic power station cannot be solved.
Disclosure of Invention
The invention provides a photovoltaic panel health state intelligent detection and analysis system and method, aiming at solving the problems of insufficient inspection precision, poor real-time performance, low efficiency and the like in a photovoltaic power station in the prior art.
The technical scheme provided by the invention is as follows:
an intelligent photovoltaic panel health status detection and analysis system, the system comprising:
preferably, the data collector and the monitoring server are in communication connection;
the data acquisition unit is used for acquiring photovoltaic panel array data;
the monitoring server is used for deploying a Fast R-CNN algorithm to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and determining an evaluation result;
wherein, the data collection station includes: set up the infrared/visible light camera of high definition on unmanned aerial vehicle, the camera is used for acquireing the image data of photovoltaic board array.
Preferably, the data collector further comprises:
the ultrasonic probe is arranged on the photovoltaic cleaning robot and used for collecting the internal structure of the photovoltaic panel; the meteorological environment monitor is used for acquiring field environment parameters of the photovoltaic power station; the data acquisition card is used for acquiring operation parameters, environment monitoring parameters and ultrasonic image parameters in the photovoltaic array;
the ultrasonic probe, the meteorological environment monitor and the data acquisition card are respectively in communication connection with the monitoring server.
Preferably, the Fast R-CNN algorithm comprises: the system comprises a deep convolutional network unit, an interested region pooling unit, a preselected region classification unit and a mark frame regression unit;
the deep convolutional network unit is used for inputting the photovoltaic panel array data into the deep convolutional network unit to obtain feature maps with different sizes;
the region of interest pooling unit is used for generating a feature expression with fixed dimension for the feature map;
the preselected region classifying unit is used for classifying the image feature expression and identifying the abnormal state of the photovoltaic panel;
the marking frame regression unit is used for positioning the position of the photovoltaic panel in the form of a marking frame in the photovoltaic panel array data;
wherein, the characteristic graph is used for depicting the image characteristics.
Based on the same inventive concept, the present invention also provides an analysis method based on the intelligent monitoring system for health status of photovoltaic panels as claimed in claims 1-3, which is characterized by comprising:
acquiring photovoltaic panel array data by using a data acquisition unit;
deploying a Fast R-CNN algorithm through a monitoring server to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and obtaining an evaluation result;
wherein, the data collection station includes: the high-definition infrared/visible light camera is arranged on the unmanned aerial vehicle;
the photovoltaic panel array data comprises: the image data is acquired through a high-definition infrared/visible light camera arranged on the unmanned aerial vehicle.
Preferably, the acquiring of the photovoltaic panel array data by using the data collector further includes:
collecting an ultrasonic image of the internal structure of the photovoltaic panel by using an ultrasonic probe arranged on the photovoltaic cleaning robot;
collecting the field environment parameters of the photovoltaic power station by using a meteorological environment monitor;
and collecting the operating parameters, the environmental monitoring parameters and the ultrasonic image parameters in the photovoltaic array by using a data acquisition card.
Preferably, the step of deploying Fast R-CNN algorithm by the monitoring server to analyze the photovoltaic panel array data, evaluating the health status of the photovoltaic panel, and obtaining an evaluation result includes:
inputting the photovoltaic panel array data shot by the high-definition infrared/visible light camera into a deep convolutional network to obtain pre-selection frame characteristic graphs with different sizes;
generating an image feature expression with fixed dimension for the pre-selected frame feature map;
classifying the image feature expression, and identifying the abnormal state of the photovoltaic panel shot by the high-definition infrared/visible light camera;
positioning the position of the photovoltaic panel in the form of a marking frame for the photovoltaic panel array data;
the pre-selection frame feature map comprises a plurality of pre-selection frames used for describing image features.
Preferably, the photovoltaic panel array data shot by the high-definition infrared/visible light camera is input into a deep convolutional network to obtain preselected frame feature maps with different sizes, and the method includes:
and during feature extraction, inputting the photovoltaic panel array data shot by the high-definition infrared/visible light camera into a deep convolutional network, obtaining feature graphs with different sizes through convolution operation in the network, and depicting information of all targets to be detected.
Preferably, the generating of the image feature expression with fixed dimension for the preselected frame feature map includes:
carrying out global aggregation operation on the feature map in an abnormal state area shot by the high-definition infrared/visible light camera carried by the unmanned aerial vehicle according to the size of each preselection frame, and converting two-dimensional feature map data into one-dimensional data;
and butting the one-dimensional data with a full-connection network in a Fast R-CNN algorithm, thereby realizing the image feature expression and abstract description of the fixed dimension.
Preferably, the classifying the image feature expression includes:
an SVM classifier is used for butt joint with the output end of the full-connection network, and the generalized image feature expressions are summarized to the corresponding abnormal state labels, so that image classification is realized;
wherein the SVM classifier is pre-established based on an abnormal state type.
Preferably, the positioning of the photovoltaic panel position in the form of a marked frame in the photovoltaic panel array data includes:
training a Fast R-CNN algorithm to identify image features in the preselected frame, and generating a labeling frame according to the preselected frame information by adopting the image features;
wherein the image features approximate the true position of the abnormal state.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention provides an intelligent monitoring and analyzing system for the health state of a photovoltaic panel, which comprises: the data acquisition device and the monitoring server are in communication connection; the data acquisition unit is used for acquiring photovoltaic panel array data; the monitoring server is used for deploying a Fast R-CNN algorithm to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and determining an evaluation result; wherein, the data collection station includes: the invention provides a photovoltaic panel health state intelligent monitoring and analyzing system which is simple in structure and flexible and convenient to use, does not need manual participation in a data acquisition process, and greatly improves inspection efficiency while operation consistency is guaranteed.
2. The invention provides an intelligent monitoring and analyzing method for the health state of a photovoltaic panel, which comprises the following steps: acquiring photovoltaic panel array data by using a data acquisition unit; deploying a Fast R-CNN algorithm through a monitoring server to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and obtaining an evaluation result; wherein, the data collection station includes: the high-definition infrared/visible light camera is arranged on the unmanned aerial vehicle; the photovoltaic panel array data comprises: through the image data that set up the infrared/visible light camera of high definition on unmanned aerial vehicle acquireed, solved and patrolled and examined the technical problem that the precision is not enough, the real-time is poor, inefficiency in present photovoltaic power plant.
3. Compared with a conventional manual inspection mode, the intelligent monitoring and analyzing method for the health state of the photovoltaic panel has the advantages that unmanned aerial vehicle inspection is intelligent operation, the flying speed is flexible and adjustable, and the inspection efficiency is obviously improved.
4. The invention provides an intelligent monitoring and analyzing method for the health state of a photovoltaic panel, which intelligently analyzes the health state of the photovoltaic panel by a Fast R-CNN algorithm and effectively avoids subjective interference in the traditional manual inspection process.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a schematic flow diagram of the process of the present invention;
wherein: the system comprises an unmanned aerial vehicle charging station 1, a meteorological environment monitor 2, a high-definition camera-carrying unmanned aerial vehicle 3, a data center 4, a monitoring server 5, a photovoltaic cleaning robot 6, an ultrasonic probe 7, a photovoltaic panel 8, a photovoltaic array header box 9 and a power grid 10.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the system schematic diagram of the invention is shown, and the invention is composed of an unmanned aerial vehicle, an unmanned aerial vehicle charging station, a high-definition infrared/visible light camera, a photovoltaic panel, a photovoltaic cleaning robot, an ultrasonic probe, a meteorological environment monitor, a photovoltaic array combiner box, a data center, a power grid, a monitoring server and a related deep learning algorithm module.
Example 1
The invention provides an intelligent monitoring and analyzing system for the health state of a photovoltaic panel, which utilizes a cross-media comprehensive analysis method to collect images outside and ultrasonic data inside the photovoltaic panel;
recording working parameters and environmental parameters of the photovoltaic power station during operation, and identifying abnormal states such as dirt, damage, falling and hidden crack;
the degree of influence of these abnormal states on the health condition of the photovoltaic panel is analyzed by integrating working parameters and environmental parameters, and the operation and maintenance personnel are assisted to carry out work, which specifically comprises the following steps:
the data acquisition device and the monitoring server are in communication connection;
the data acquisition unit is used for acquiring photovoltaic panel array data;
the monitoring server is used for deploying a Fast R-CNN algorithm to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and determining an evaluation result;
the data center is arranged between the data acquisition device and the monitoring server and used for summarizing the acquired image data, the environmental parameter data and the operation parameter data;
specifically, to the current photovoltaic power plant problem that the precision is not enough of patrolling and examining:
the unmanned aerial vehicle is arranged above the photovoltaic panel array to cruise, and the high-definition infrared/visible light camera is mounted to shoot the photovoltaic panel image in a proximity mode; carrying an ultrasonic probe by a photovoltaic cleaning robot to scan the internal structure of the photovoltaic panel; a meteorological environment monitor is used for collecting environmental parameters such as field solar radiation quantity, temperature, humidity, wind speed, wind direction and the like of the photovoltaic power station; the collection function collection box is used for extracting operating parameters such as current, voltage and power in the working process of the photovoltaic panel array. Each module establishes a small-sized Internet of things in a wireless transmission mode; the Internet of things is established through a 5G network; transmitting collected data such as images, environmental parameters, operation parameters and the like back to a data center in real time through the Internet of things; the data center is matched and summarized according to time sequence and type, and then transmitted to a monitoring server for analysis; and the present invention proposes: high-definition and high-precision equipment is used for collecting internal and external images and other parameters of the photovoltaic panel, so that the data collection precision is ensured;
the problem that the current photovoltaic power station inspection real-time performance is insufficient is solved:
the unmanned aerial vehicle related to the invention charges/stands by in the unmanned aerial vehicle charging station; the photovoltaic cleaning robot is provided with solar charging equipment and is deployed on the photovoltaic panel array; a meteorological environment monitor and a collection box with a collection function collect target parameters at a fixed frequency, and extract image parameters; transmitting the data to a data center in real time in a wireless transmission mode for gathering and integrating; the unmanned aerial vehicle and the cleaning robot are matched with the meteorological environment monitor and the header box with the collection function, respond to commands in real time and patrol the specific photovoltaic panel array or the comprehensive patrol;
to the problem that current photovoltaic power plant patrols and examines inefficiency:
the unmanned aerial vehicle provided by the invention has an autonomous navigation function; after the unmanned aerial vehicle operator controls shooting of the panoramic image of the photovoltaic power station, the unmanned aerial vehicle system can automatically establish a photovoltaic power station model; the cruise line, the flight speed, the flight height and the image acquisition frequency are automatically planned; automatically monitoring the electricity usage condition and determining the return charging time; the unmanned aerial vehicle charging station provides a charging/standby place for the unmanned aerial vehicle, and the charging station is provided with a photovoltaic panel for supplying power to the charging/standby place; the electric quantity use condition of the unmanned aerial vehicle is automatically monitored; compared with a conventional manual inspection mode, the unmanned aerial vehicle inspection is intelligent, the flying speed is flexible and adjustable, and the inspection efficiency is obviously improved;
aiming at the problem that the current photovoltaic panel is detected by electroluminescence to cause hidden cracking operation to be complex:
according to the invention, the ultrasonic probe is arranged on the photovoltaic cleaning robot; the internal structure of the photovoltaic panel can be detected while the cleaning task is executed; and generating an ultrasonic image for Fast R-CNN algorithm training. The invention has simple structure and flexible and convenient use, does not need manual participation in the data acquisition process, and greatly improves the inspection efficiency while ensuring the operation consistency.
In the embodiment, the Fast R-CNN algorithm is used for identifying the abnormal states of dirt, damage, falling and subfissure of the photovoltaic panel; preprocessing the collected visible light image, infrared image and ultrasonic image;
marking abnormal state areas of the photovoltaic panel, such as dirt, damage, falling, hidden cracks and the like, in the image, and preparing a data set by using the marked image;
and identifying the image characteristics of the abnormal state of the photovoltaic panel by training a Fast R-CNN algorithm, and displaying the image characteristics in the form of a labeling box in the image.
The algorithm comprises a deep convolutional network unit, an interested region pooling unit, a preselected region classification unit and a labeling frame regression unit;
the deep convolutional network unit is used for inputting the photovoltaic panel array data into the deep convolutional network unit to obtain feature maps with different sizes;
the region of interest pooling unit is used for generating a feature expression with fixed dimension for the feature map;
the preselected region classifying unit is used for classifying the image feature expression and identifying the abnormal state of the photovoltaic panel;
the marking frame regression unit is used for positioning the position of the photovoltaic panel in the form of a marking frame in the photovoltaic panel array data;
wherein, the characteristic graph is used for depicting the image characteristics.
Example 2
The invention also provides an analysis method of the intelligent photovoltaic panel health state monitoring system, and the flow of the method is shown in figure 2.
Step 1: acquiring photovoltaic panel array data by using a data acquisition unit;
step 2: deploying a Fast R-CNN algorithm through a monitoring server to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and obtaining an evaluation result;
wherein, the data collection station includes: the device comprises a high-definition infrared/visible light camera arranged on the unmanned aerial vehicle and an ultrasonic probe arranged on the photovoltaic cleaning robot and used for collecting the internal structure of a photovoltaic panel; the meteorological environment monitor is used for acquiring field environment parameters of the photovoltaic power station; the data acquisition card is used for acquiring operation parameters, environment monitoring parameters and ultrasonic image parameters in the photovoltaic array;
the ultrasonic probe, the meteorological environment monitor and the data acquisition card are respectively in communication connection with the monitoring server;
the photovoltaic panel array data comprises: the image data is acquired through a high-definition infrared/visible light camera arranged on the unmanned aerial vehicle.
The method comprises the following steps of analyzing the photovoltaic panel array data by deploying a Fast R-CNN algorithm through a monitoring server, evaluating the health state of the photovoltaic panel, and obtaining an evaluation result, wherein the evaluation result comprises the following steps:
inputting the photovoltaic panel array data shot by the high-definition infrared/visible light camera into a deep convolutional network to obtain pre-selection frame characteristic graphs with different sizes;
generating an image feature expression with fixed dimension for the pre-selected frame feature map;
classifying the image feature expression, and identifying the abnormal state of the photovoltaic panel shot by the high-definition infrared/visible light camera;
positioning the position of the photovoltaic panel in the form of a marking frame for the photovoltaic panel array data;
the pre-selection frame feature map comprises a plurality of pre-selection frames for depicting image features;
the Fast R-CNN algorithm comprises a deep convolutional network unit, an interested region pooling unit, a preselected region classification unit and a labeling frame regression unit;
the deep convolutional network unit is used for inputting the photovoltaic panel array data into the deep convolutional network unit to obtain feature maps with different sizes;
the region of interest pooling unit is used for generating a feature expression with fixed dimension for the feature map;
the preselected region classifying unit is used for classifying the image feature expression and identifying the abnormal state of the photovoltaic panel;
the marking frame regression unit is used for positioning the position of the photovoltaic panel in the form of a marking frame in the photovoltaic panel array data;
wherein, the model is automatically adjusted according to the feedback condition of the loss function during the training period, thereby having the capability of identifying the image characteristics, and the image classification task selects LclsThe loss function is expressed as follows:
Lcls(p,u)＝-log pu (1)
p represents a label of the type of a labeling frame predicted by the model (the type of the labeling frame is abnormal conditions such as 'hidden crack', 'falling', 'dirty', 'damaged'), and the like in the invention); u represents a category label of the label box; puRepresenting the output of the corresponding category u of the network in the deep convolutional network;
when the characteristics are extracted, inputting the photovoltaic panel array data shot by the high-definition infrared/visible light camera into a deep convolutional network, obtaining characteristic diagrams with different sizes through convolution operation in the network, and depicting information of all targets to be detected;
generating a fixed-dimension image feature representation for a preselected frame feature map, comprising: carrying out global aggregation operation on the feature map in an abnormal state area shot by the high-definition infrared/visible light camera carried by the unmanned aerial vehicle according to the size of each preselection frame, and converting two-dimensional feature map data into one-dimensional data;
docking the one-dimensional data with a full-connection network in a Fast R-CNN algorithm, thereby realizing the image feature expression and abstract description of the fixed dimension;
classifying the image feature representation, including: an SVM classifier is used for butt joint with the output end of the full-connection network, and the generalized image feature expressions are summarized to the corresponding abnormal state labels, so that image classification is realized;
the SVM classifier is established in advance based on the abnormal state type;
finally, performing label box regression, including: training a Fast R-CNN algorithm to identify image features in the preselected frame, and generating a labeling frame according to the preselected frame information by adopting the image features;
the regression is to reduce the difference between the predicted bounding box and the real bounding box, so that the bounding box predicted by the algorithm is infinitely close to the real boundary, and the algorithm has the capability of detecting the position information of the target;
train with LlocThe loss function is an evaluation index, and the loss function selected by the position analysis task is expressed as follows:
v represents a pre-selection box corresponding to the category u; v ═ v (v)x,vy,vw,vh) The coordinates of the center point of the preselected frame and the width and height dimensions of the preselected frame are expressed as follows:
here (P)x,Py,Pw,Ph) And (G)x,Gy,Gw,Gh) Respectively representing the coordinates of the central point of the real boundary frame of the prediction frame and the width and height dimensions of the preselected frame; t represents the predicted regression objective,
tuthe method is a conversion relation between a target actual position v and a predicted position, smooths is an activation function and is used for introducing nonlinear features, and therefore the algorithm is easier to fit a complex process.
The scheme provided by the invention can effectively solve the practical problems of insufficient inspection precision, poor real-time performance, low efficiency and the like in the conventional photovoltaic power station.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that the above-mentioned embodiments are only preferred embodiments of the present invention, and are not intended to limit the present invention, and those skilled in the art will appreciate that various modifications and changes can be made to the present invention. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of the present invention is included in the scope of the claims of the present invention filed as filed.
Claims (10)
1. An intelligent photovoltaic panel health monitoring and analysis system, the system comprising: the data acquisition device and the monitoring server are in communication connection;
the data acquisition unit is used for acquiring photovoltaic panel array data;
the monitoring server is used for deploying a Fast R-CNN algorithm to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and determining an evaluation result;
wherein, the data collection station includes: set up the infrared/visible light camera of high definition on unmanned aerial vehicle, the camera is used for acquireing the image data of photovoltaic board array.
2. The intelligent monitoring and analysis system of claim 1, wherein the data collector further comprises:
the ultrasonic probe is arranged on the photovoltaic cleaning robot and used for collecting the internal structure of the photovoltaic panel; the meteorological environment monitor is used for acquiring field environment parameters of the photovoltaic power station; the data acquisition card is used for acquiring operation parameters, environment monitoring parameters and ultrasonic image parameters in the photovoltaic array;
the ultrasonic probe, the meteorological environment monitor and the data acquisition card are respectively in communication connection with the monitoring server.
3. The intelligent monitoring and analysis system of claim 1, wherein the Fast R-CNN algorithm comprises: the system comprises a deep convolutional network unit, an interested region pooling unit, a preselected region classification unit and a mark frame regression unit;
the deep convolutional network unit is used for inputting the photovoltaic panel array data into the deep convolutional network unit to obtain feature maps with different sizes;
the region of interest pooling unit is used for generating a feature expression with fixed dimension for the feature map;
the preselected region classifying unit is used for classifying the image feature expression and identifying the abnormal state of the photovoltaic panel;
the marking frame regression unit is used for positioning the position of the photovoltaic panel in the form of a marking frame in the photovoltaic panel array data;
wherein, the characteristic graph is used for depicting the image characteristics.
4. An analysis method based on the intelligent photovoltaic panel health state monitoring system according to claims 1-3, characterized by comprising the following steps:
acquiring photovoltaic panel array data by using a data acquisition unit;
deploying a Fast R-CNN algorithm through a monitoring server to analyze the photovoltaic panel array data, evaluating the health state of the photovoltaic panel and obtaining an evaluation result;
wherein, the data collection station includes: the high-definition infrared/visible light camera is arranged on the unmanned aerial vehicle;
the photovoltaic panel array data comprises: the image data is acquired through a high-definition infrared/visible light camera arranged on the unmanned aerial vehicle.
5. The method of claim 4, wherein the acquiring photovoltaic panel array data with the data collector further comprises:
collecting an ultrasonic image of the internal structure of the photovoltaic panel by using an ultrasonic probe arranged on the photovoltaic cleaning robot;
collecting the field environment parameters of the photovoltaic power station by using a meteorological environment monitor;
and collecting the operating parameters, the environmental monitoring parameters and the ultrasonic image parameters in the photovoltaic array by using a data acquisition card.
6. The method according to claim 4, wherein the monitoring server deployment Fast R-CNN algorithm analyzes the photovoltaic panel array data, evaluates photovoltaic panel health status, and obtains evaluation results, and comprises:
inputting the photovoltaic panel array data shot by the high-definition infrared/visible light camera into a deep convolutional network to obtain pre-selection frame characteristic graphs with different sizes;
generating an image feature expression with fixed dimension for the pre-selected frame feature map;
classifying the image feature expression, and identifying the abnormal state of the photovoltaic panel shot by the high-definition infrared/visible light camera;
positioning the photovoltaic panel position in the form of a marking frame for the photovoltaic panel array data;
the pre-selection frame feature map comprises a plurality of pre-selection frames used for describing image features.
7. The method of claim 6, wherein inputting the photovoltaic panel array data captured by the high definition infrared/visible camera into a deep convolutional network to obtain pre-selected box feature maps of different sizes comprises:
and during feature extraction, inputting the photovoltaic panel array data shot by the high-definition infrared/visible light camera into a deep convolutional network, obtaining feature graphs with different sizes through convolution operation in the network, and depicting information of all targets to be detected.
8. The method of claim 6, wherein generating a fixed-dimension representation of image features for a preselected frame feature map comprises:
carrying out global aggregation operation on the feature map in an abnormal state area shot by the high-definition infrared/visible light camera carried by the unmanned aerial vehicle according to the size of each preselection frame, and converting two-dimensional feature map data into one-dimensional data;
and butting the one-dimensional data with a full-connection network in a Fast R-CNN algorithm, thereby realizing the image feature expression and abstract description of the fixed dimension.
9. The method of claim 8, wherein the classifying the image feature representation comprises:
an SVM classifier is used for butt joint with the output end of the full-connection network, and the generalized image feature expressions are summarized to the corresponding abnormal state labels, so that image classification is realized;
wherein the SVM classifier is pre-established based on an abnormal state type.
10. The method of claim 6, wherein the photovoltaic panel array data locates photovoltaic panel locations in the form of labeled boxes comprising:
training a Fast R-CNN algorithm to identify image features in the preselected frame, and generating a labeling frame according to the preselected frame information by adopting the image features;
wherein the image features approximate the true position of the abnormal state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910846118.9A CN112455676A (en) | 2019-09-09 | 2019-09-09 | Intelligent monitoring and analyzing system and method for health state of photovoltaic panel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910846118.9A CN112455676A (en) | 2019-09-09 | 2019-09-09 | Intelligent monitoring and analyzing system and method for health state of photovoltaic panel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112455676A true CN112455676A (en) | 2021-03-09 |
Family
ID=74807000
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910846118.9A Pending CN112455676A (en) | 2019-09-09 | 2019-09-09 | Intelligent monitoring and analyzing system and method for health state of photovoltaic panel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112455676A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113436176A (en) * | 2021-07-01 | 2021-09-24 | 华风数据(深圳)有限公司 | Photovoltaic panel surface fault detection method based on SSD and DSSD algorithm under visible light |
| CN113485432A (en) * | 2021-07-26 | 2021-10-08 | 西安热工研究院有限公司 | Photovoltaic power station electroluminescence intelligent diagnosis system and method based on unmanned aerial vehicle |
| CN116432089A (en) * | 2023-05-15 | 2023-07-14 | 厦门星拉科技有限公司 | Electric power internet of things inspection system and method |
| CN117853827A (en) * | 2024-03-07 | 2024-04-09 | 安徽省大气探测技术保障中心 | Sampling pump working state operation monitoring system and method for atmospheric chamber gas monitoring |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011043416A (en) * | 2009-08-21 | 2011-03-03 | Toyota Central R&D Labs Inc | Ultrasonic inspection system |
| CN106160659A (en) * | 2016-08-24 | 2016-11-23 | 河海大学常州校区 | A kind of photovoltaic plant region orientation method for diagnosing faults |
| CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
| CN107111313A (en) * | 2014-12-17 | 2017-08-29 | Abb瑞士股份有限公司 | Solar panel is checked with unmanned vehicle |
| CN108418548A (en) * | 2018-02-11 | 2018-08-17 | 同济大学 | Photovoltaic panel omnibearing integrated monitors system |
| CN108418550A (en) * | 2018-03-16 | 2018-08-17 | 南京云思创智信息科技有限公司 | The defects detection and localization method of cluster solar panels |
| CN109724547A (en) * | 2019-02-28 | 2019-05-07 | 泸州金能移动能源科技有限公司 | A kind of photovoltaic module dislocation on-line measuring device |
| CN110008927A (en) * | 2019-04-15 | 2019-07-12 | 河南大华安防科技股份有限公司 | One kind automating determination method based on the improved alert of deep learning model Fast-Rcnn |
-
2019
- 2019-09-09 CN CN201910846118.9A patent/CN112455676A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011043416A (en) * | 2009-08-21 | 2011-03-03 | Toyota Central R&D Labs Inc | Ultrasonic inspection system |
| CN107111313A (en) * | 2014-12-17 | 2017-08-29 | Abb瑞士股份有限公司 | Solar panel is checked with unmanned vehicle |
| CN106160659A (en) * | 2016-08-24 | 2016-11-23 | 河海大学常州校区 | A kind of photovoltaic plant region orientation method for diagnosing faults |
| CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
| CN108418548A (en) * | 2018-02-11 | 2018-08-17 | 同济大学 | Photovoltaic panel omnibearing integrated monitors system |
| CN108418550A (en) * | 2018-03-16 | 2018-08-17 | 南京云思创智信息科技有限公司 | The defects detection and localization method of cluster solar panels |
| CN109724547A (en) * | 2019-02-28 | 2019-05-07 | 泸州金能移动能源科技有限公司 | A kind of photovoltaic module dislocation on-line measuring device |
| CN110008927A (en) * | 2019-04-15 | 2019-07-12 | 河南大华安防科技股份有限公司 | One kind automating determination method based on the improved alert of deep learning model Fast-Rcnn |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113436176A (en) * | 2021-07-01 | 2021-09-24 | 华风数据(深圳)有限公司 | Photovoltaic panel surface fault detection method based on SSD and DSSD algorithm under visible light |
| CN113485432A (en) * | 2021-07-26 | 2021-10-08 | 西安热工研究院有限公司 | Photovoltaic power station electroluminescence intelligent diagnosis system and method based on unmanned aerial vehicle |
| CN116432089A (en) * | 2023-05-15 | 2023-07-14 | 厦门星拉科技有限公司 | Electric power internet of things inspection system and method |
| CN117853827A (en) * | 2024-03-07 | 2024-04-09 | 安徽省大气探测技术保障中心 | Sampling pump working state operation monitoring system and method for atmospheric chamber gas monitoring |
| CN117853827B (en) * | 2024-03-07 | 2024-05-14 | 安徽省大气探测技术保障中心 | Sampling pump working state operation monitoring system and method for atmospheric chamber gas monitoring |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112455676A (en) | Intelligent monitoring and analyzing system and method for health state of photovoltaic panel | |
| CN110261394B (en) | Online real-time diagnosis system and method for damage of fan blade | |
| CN108037133B (en) | Intelligent electric power equipment defect identification method and system based on unmanned aerial vehicle inspection image | |
| CN112633535A (en) | Photovoltaic power station intelligent inspection method and system based on unmanned aerial vehicle image | |
| CN110033453A (en) | Based on the power transmission and transformation line insulator Aerial Images fault detection method for improving YOLOv3 | |
| CN111198004A (en) | Electric power inspection information acquisition system based on unmanned aerial vehicle | |
| CN109060826B (en) | Wind-powered electricity generation blade detection device that does not shut down | |
| CN113205116B (en) | Automatic extraction and track planning method for inspection shooting target point of unmanned aerial vehicle of power transmission line | |
| CN112360699A (en) | Intelligent inspection and diagnosis analysis method for blades of full-automatic wind generating set | |
| CN108956640A (en) | Vehicle-mounted detection apparatus and detection method suitable for distribution line inspection | |
| CN111539355A (en) | Photovoltaic panel foreign matter detection system and detection method based on deep neural network | |
| CN113160184B (en) | Unmanned aerial vehicle intelligent inspection cable surface defect detection method based on deep learning | |
| CN116231504A (en) | Remote intelligent inspection method, device and system for booster station | |
| CN111244822A (en) | Fixed-wing unmanned aerial vehicle line patrol method, system and device in complex geographic environment | |
| CN114944816B (en) | Intelligent inspection system applied to photovoltaic module | |
| Herraiz et al. | Optimal productivity in solar power plants based on machine learning and engineering management | |
| CN114648708A (en) | State detection method and device of photovoltaic module, storage medium and electronic equipment | |
| CN114663672B (en) | Method and system for detecting corrosion of steel member of power transmission line tower | |
| CN112329584A (en) | Method, system and equipment for automatically identifying foreign matters in power grid based on machine vision | |
| CN114895712A (en) | Large-range inspection method and system for photovoltaic power generation equipment | |
| CN108180871B (en) | method for quantitatively evaluating composite insulator surface pulverization roughness | |
| CN117498225A (en) | Unmanned aerial vehicle intelligent power line inspection system | |
| Liu et al. | Cloud-edge-end cooperative detection of wind turbine blade surface damage based on lightweight deep learning network | |
| Sha et al. | Design of patrol monitoring and control system for hot spot of solar photovoltaic module | |
| CN115622231A (en) | Photovoltaic power generation station intelligence operation and maintenance 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210309 |
|
| RJ01 | Rejection of invention patent application after publication |