CN116191680B - Monitoring management system applied to photovoltaic power generation - Google Patents
Monitoring management system applied to photovoltaic power generation Download PDFInfo
- Publication number
- CN116191680B CN116191680B CN202310455131.8A CN202310455131A CN116191680B CN 116191680 B CN116191680 B CN 116191680B CN 202310455131 A CN202310455131 A CN 202310455131A CN 116191680 B CN116191680 B CN 116191680B
- Authority
- CN
- China
- Prior art keywords
- module
- photovoltaic
- image
- photovoltaic module
- contour
- 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.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 66
- 238000010248 power generation Methods 0.000 title claims abstract description 31
- 230000005856 abnormality Effects 0.000 claims abstract description 60
- 230000002159 abnormal effect Effects 0.000 claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 238000003745 diagnosis Methods 0.000 claims abstract description 10
- 238000004891 communication Methods 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims description 52
- 241000270295 Serpentes Species 0.000 claims description 20
- 238000003062 neural network model Methods 0.000 claims description 19
- 238000000034 method Methods 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 9
- 230000007613 environmental effect Effects 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 7
- 238000012423 maintenance Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 7
- 210000002569 neuron Anatomy 0.000 claims description 7
- 238000003706 image smoothing Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 239000011521 glass Substances 0.000 claims description 3
- 239000000463 material Substances 0.000 abstract description 5
- 230000000694 effects Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 7
- 210000004027 cell Anatomy 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 210000003608 fece Anatomy 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- 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
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B3/00—Audible signalling systems; Audible personal calling systems
- G08B3/10—Audible signalling systems; Audible personal calling systems using electric transmission; using electromagnetic transmission
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- 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
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- 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)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Mathematical Physics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Electromagnetism (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Human Computer Interaction (AREA)
- Photovoltaic Devices (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention discloses a monitoring management system applied to photovoltaic power generation, which comprises a monitoring terminal module, a cloud server and a management terminal module; the monitoring terminal module comprises a data acquisition module and a video monitoring module; the cloud server comprises a data acquisition module, an analysis and diagnosis module, a communication module and a database; the management terminal module comprises an information inquiry module, a state display module, a control scheduling module and an abnormality alarm module. The invention can realize collection, storage and analysis processing of the operation parameters and real-time state data of the photovoltaic module, thereby realizing the prediction, fault display and alarm prompt of the operation parameters of the photovoltaic power generation system, providing comprehensive monitoring and management functions for users, detecting and identifying the phenomena of foreign matter shielding, component damage or material falling and the like on the surface of the photovoltaic module, detecting and identifying abnormal operation data, and effectively improving the operation safety of the photovoltaic power generation system.
Description
Technical Field
The invention relates to the technical field of photovoltaic power generation, in particular to a monitoring management system applied to photovoltaic power generation.
Background
At present, the global energy problem is increasingly prominent, the traditional fuel energy is being reduced every day, and the development and utilization of green and environment-friendly energy are more important. Photovoltaic power generation is used as novel renewable energy, has the advantages of safety, reliability, cleanness, no pollution, reproducibility and the like, and governments of various countries are also under the policy of supporting the photovoltaic industry. Photovoltaic power generation is a technology for directly converting light energy into electric energy by utilizing the photovoltaic effect of a semiconductor interface, and mainly comprises three parts of a solar panel (component), a controller and an inverter, wherein the main parts comprise electronic components. The solar cells are packaged and protected after being connected in series to form a large-area solar cell module, and then the solar cell module is matched with components such as a power controller and the like to form the photovoltaic power generation device.
However, in order to save land cost and increase illumination time, the existing large-scale photovoltaic power plant often selects a construction address in a remote area or even a high-altitude area, so that the severe conditions bring inconvenience to maintenance of photovoltaic power generation equipment. Meanwhile, as the photovoltaic cell panel is exposed outdoors throughout the year, the phenomena of dust and bird droppings shielding, the problems of damage of the photovoltaic module or falling of the board and the like occur frequently, if the problems cannot be eliminated in time, the local temperature rise of the photovoltaic panel or the hot spot effect occurs, so that the power generation efficiency is reduced, even the board is short-circuited to cause the burning of the module, and in addition, the phenomena of short circuit, abnormal output power and the like are easily caused by abnormal operation in the operation process of the photovoltaic module, so that the power quality of a photovoltaic power station is improved, the normal operation of the photovoltaic power generation is ensured, and the inspection and the monitoring of the photovoltaic module are required.
Disclosure of Invention
Aiming at the problems in the related art, the invention provides a monitoring management system applied to photovoltaic power generation, so as to overcome the technical problems existing in the prior related art.
For this purpose, the invention adopts the following specific technical scheme:
a monitoring management system applied to photovoltaic power generation comprises a monitoring terminal module, a cloud server and a management terminal module;
the monitoring terminal module comprises a data acquisition module and a video monitoring module;
the data acquisition module is used for acquiring environmental data and operation data of the photovoltaic power station;
the video monitoring module is used for carrying out real-time video monitoring on the field condition of the photovoltaic power station;
the cloud server comprises a data acquisition module, an analysis and diagnosis module, a communication module and a database;
the data acquisition module is used for acquiring environmental data, operation data and on-site real-time video monitoring data of the photovoltaic power station;
the analysis and diagnosis module is used for carrying out abnormality detection on the photovoltaic module of the photovoltaic power station by utilizing an image recognition technology and also used for carrying out analysis and detection on the operation state of the photovoltaic module by utilizing the acquired operation data;
the communication module is used for realizing wireless communication connection between the cloud server and the monitoring terminal module and between the cloud server and the management terminal module;
The database is used for storing and calling data generated in the running process of the system;
the management terminal module comprises an information inquiry module, a state display module, a control scheduling module and an abnormality alarm module;
the information inquiry module is used for inquiring the environmental data, the operation data and the on-site real-time video monitoring data of the photovoltaic power station by an administrator;
the state display module is used for displaying environment data, operation data and on-site real-time video monitoring data of the photovoltaic power station;
the control scheduling module is used for executing a corresponding control scheduling plan according to the collected operation data and the detected abnormal data;
the abnormality alarming module is used for alarming when the operation state of the photovoltaic power station is abnormal, and sending a maintenance request notice to a manager.
Further, the data acquisition module comprises a temperature and humidity acquisition module, a current acquisition module, a voltage acquisition module and a power acquisition module;
the temperature and humidity acquisition module is used for acquiring real-time temperature and humidity of the photovoltaic power station when the on-site photovoltaic module operates;
the current acquisition module is used for acquiring real-time running current of the photovoltaic module and the photovoltaic grid-connected inverter;
the voltage acquisition module is used for acquiring real-time operation voltage of the photovoltaic module and the photovoltaic grid-connected inverter;
The power acquisition module is used for acquiring real-time operation power of the photovoltaic module and the photovoltaic grid-connected inverter.
Further, the analysis and diagnosis module comprises a photovoltaic module abnormality detection module and an operation state abnormality detection module;
the photovoltaic module abnormality detection module is used for detecting abnormality of the photovoltaic module in the photovoltaic power station by utilizing an image recognition technology, so that abnormality detection of the state of the photovoltaic module is realized;
the operation state abnormality detection module is used for analyzing and detecting the operation state of the photovoltaic module by utilizing the improved BP wavelet neural network model and combining the acquired operation data so as to realize abnormality detection of the operation state of the photovoltaic module;
the photovoltaic module abnormality detection module comprises a photovoltaic module identification module and a photovoltaic module abnormality identification module;
the photovoltaic module identification module is used for identifying the photovoltaic module in the field real-time video monitoring data by utilizing the improved GVF Snake model;
the photovoltaic module abnormality recognition module is used for recognizing and classifying foreign matter shielding or module breakage conditions of the photovoltaic module by utilizing a probability Hough transform algorithm and combining a contour extraction algorithm.
Further, the photovoltaic module identification module comprises a sequence image acquisition module, an image preprocessing module, a GVF image generation module, an initial convergence position matching module, an iteration convergence module, an initial convergence contour position matching module, a contour point prediction module, a Snake iteration module and an identification result output module;
The sequence image acquisition module is used for acquiring sequence images of the live real-time monitoring video;
the image preprocessing module is used for carrying out image enhancement processing when the sequence image has a reflection phenomenon and also is used for carrying out smoothing processing on the sequence image;
the GVF image generation module is used for calculating by utilizing the improved GVF model and iteratively generating an improved image GVF;
the initial convergence position matching module is used for clicking a set key point position near the edge of the image in a mouse clicking mode along the edge of the photovoltaic module string in the first sequence image, and the obtained convergence profile is used for predicting a second layer profile point and obtaining an initial convergence position through matching;
the iteration convergence module is used for iterating by taking the contour of the initial convergence position as the initial contour of the Snake;
the initial convergence contour position matching module is used for taking the position of the first sequence image as the predicted position of the second sequence image and also as the matched initial position, obtaining the initial convergence contour position after matching, judging whether the initial convergence contour position exceeds the bottom layer, if so, realizing the identification of the photovoltaic module string in the sequence image, and if not, carrying out block matching;
the contour point prediction module is used for predicting the contour point of the next layer by utilizing the convergence result of the first two layers and performing block matching by taking the predicted position as an initial position;
The Snake iteration module is used for carrying out iteration convergence by utilizing the Snake with three-dimensional constraint, carrying out layer-by-layer judgment on the layer number again and again, and carrying out subsequent prediction and matching, so as to circulate until the identification of all sequence images is realized;
the identification result output module is used for outputting identification results of the photovoltaic module strings in all the sequence images and numbering all the photovoltaic module strings.
Further, the image preprocessing module comprises a reflection judgment module, an image enhancement processing module and an image smoothing processing module;
the reflection judging module is used for judging whether the sequence image has a reflection phenomenon or not;
the image enhancement processing module is used for carrying out image enhancement processing when the light reflection phenomenon occurs in the column images;
the image smoothing processing module is used for smoothing the sequence images, enhancing edges and removing noise.
Further, the expression of the improved GVF model is:
where ε represents the energy function, f represents the edge map of the processed image, (x, y) represents the coordinates of the pixels, D represents a symmetric and semi-positive matrix, μ represents the weighting coefficient, V represents the vector field, f represents the gradient field of f, u represents the projection of the force field in the x-direction, V represents the projection of the force field in the y-direction, V represents the gradient operator of the image solution, dx represents the convolution kernel in the x-direction, dy represents the convolution kernel in the y-direction, and T represents the transpose.
Further, the photovoltaic module abnormality recognition module comprises a photovoltaic module image acquisition module, an abnormality contour extraction module and an abnormality state recognition module;
the photovoltaic module image acquisition module is used for intercepting a photovoltaic module area in the minimum circumscribed rectangle of the module according to the identification result of the photovoltaic module to obtain a single photovoltaic module image;
the abnormal image acquisition module is used for eliminating a background plate and grid lines of the single photovoltaic module image based on morphology, eliminating a frame of the single photovoltaic module image based on probability Hough transform algorithm, and obtaining an abnormal image;
the abnormal contour extraction module is used for extracting the contour of the abnormal image by utilizing an eight-domain traversal algorithm to obtain an extracted contour map;
the abnormal state identification module is used for identifying and classifying the abnormal state of the photovoltaic module based on the extracted contour map;
the abnormal state identification module identifies and classifies the abnormal state of the photovoltaic module based on the extracted contour map, and comprises the following steps:
judging whether the effective contour is extracted, if not, indicating that no abnormality exists, if yes, judging whether a plurality of contours exist, if not, judging whether to execute color features in the contours, and if yes, executing the next step;
After the screening threshold is reduced, continuously judging whether radial areas appear on the multiple outlines, if so, judging that the photovoltaic module glass is damaged, and if not, judging that the multiple foreign matters are shielded;
judging whether the color features in the outline are non-white areas, if so, judging that the outline is shielded by the foreign matter, if not, judging that the outline has regularity, if so, judging that the EVA falls off, and if not, judging that the outline is shielded by the foreign matter.
Further, when it is determined that the foreign matter is blocked, the method includes the steps of:
extracting the minimum circumscribed rectangle of the extracted contour image to obtain the minimum circumscribed rectangle of all contours, and simultaneously calculating the length, width, length-width ratio and area information of all the minimum circumscribed rectangle;
screening and removing all minimum circumscribed rectangles based on a preset contour screening condition, screening out areas belonging to foreign matter shielding, and obtaining the contour of the foreign matter shielding and the position information of the shielding areas;
wherein, the profile screening conditions are:
condition 1: width is greater than or equal to LTH & Length is less than or equal to HTH & Length is greater than or equal to Width
Condition 2: a is less than or equal to length/Width is less than or equal to b
In the formula, width represents the minimum circumscribed rectangle Width, length represents the minimum circumscribed rectangle length, LTH represents the low threshold value size, HTH represents the high threshold value size, a and b represent range parameters, and when the minimum circumscribed rectangle meets the condition 1 and the condition 2 at the same time, the minimum circumscribed rectangle is considered to be a foreign matter shielding area.
Further, the operation state anomaly detection module analyzes and detects the operation state of the photovoltaic module by utilizing the improved BP wavelet neural network model and combining the acquired operation data, and the implementation of anomaly detection of the operation state of the photovoltaic module comprises the following steps:
constructing an improved BP wavelet neural network model, and training the model by utilizing historical operation data of a photovoltaic power station in a database;
collecting operation data of the photovoltaic module and the photovoltaic grid-connected inverter at the previous moment, and inputting the operation data into a trained improved BP wavelet neural network model to obtain predicted operation data of the photovoltaic module and the photovoltaic grid-connected inverter at the current moment;
and analyzing and comparing the operation data of the photovoltaic module and the photovoltaic grid-connected inverter obtained by detection at the current moment with the predicted operation data, and judging that the operation state of the photovoltaic module at the current moment is abnormal when the difference value between the operation data and the predicted operation data exceeds a preset threshold value.
Further, constructing an improved BP wavelet neural network model comprises the following steps:
setting an input layer to a hidden layer weight value, a hidden layer neuron threshold value and a wavelet expansion translation parameter by combining wavelet type, time-frequency parameters and training sample data;
Constructing a new error function by introducing hidden layer saturation, and establishing an improved BP wavelet neural network model;
wherein the new error function is formulated as follows:
wherein P represents the total number of input samples, P represents the samples, m represents the number of nodes of the output layer, j represents each node of the output layer, l represents the number of nodes of the hidden layer, k represents each node of the hidden layer, N represents an even number, and t p Indicating the desired output, t pj Representing the expected output of each node of the output layer, t pk Representing the expected output of each node of the hidden layer, y p Representing the actual output, y pj Representing the actual output of each node of the output layer,representing the hidden layer saturation scale factor, E A Represents the A-th sample error, E B Indicating sample B error.
The beneficial effects of the invention are as follows:
1) Through the cooperation use among the monitoring terminal module, the cloud server and the management terminal module, the collection, storage and analysis processing of the operation parameters and the real-time state data of the photovoltaic module can be realized, so that the operation parameter prediction, fault display, alarm prompt and the like of the photovoltaic power generation system can be realized, and further, the comprehensive monitoring management function can be provided for a user.
2) Through being provided with photovoltaic module anomaly detection module and the unusual detection module of running state to can not only utilize image recognition technology to realize the detection and the discernment of phenomenon such as photovoltaic module surface foreign matter shelters from, the damaged or material that drops of subassembly under the unusual detection module's of photovoltaic module effect, can also utilize the unusual detection module of running state to utilize modified BP wavelet neural network model to realize the detection and the discernment to unusual operation data, and then can improve photovoltaic power generation system's operation safety effectively.
3) Through being provided with photovoltaic module identification module and photovoltaic module unusual identification module to can utilize the photovoltaic module of modified GVF Snake model in the scene real-time video monitoring data under photovoltaic module identification module's effect discernment, can be better consider the marginal information of image in the image segmentation, make the separation effect more ideal, can also utilize probability Hough transform algorithm to combine the recognition of phenomena such as contour extraction algorithm such as photovoltaic module surface foreign matter shelters from, the subassembly is damaged or the material drops under photovoltaic module unusual identification module's effect, and then can be better satisfy photovoltaic power plant's control management demand.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed 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 other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of a monitoring management system applied to photovoltaic power generation according to an embodiment of the present invention;
Fig. 2 is a block diagram of a photovoltaic module abnormality detection module applied to a monitoring management system for photovoltaic power generation according to an embodiment of the present invention;
FIG. 3 is a block diagram of a photovoltaic module identification module in a monitoring management system for photovoltaic power generation according to an embodiment of the present invention;
fig. 4 is a block diagram of a photovoltaic module abnormality recognition module applied to a monitoring management system for photovoltaic power generation according to an embodiment of the present invention.
In the figure:
1. a monitoring terminal module; 11. a data acquisition module; 111. a temperature and humidity acquisition module; 112. a current collection module; 113. a voltage acquisition module; 114. a power acquisition module; 12. a video monitoring module; 2. the cloud server; 21. a data acquisition module; 22. an analytical diagnostic module; 221. the photovoltaic module abnormality detection module; 2211. a photovoltaic module identification module; 22111. a sequential image acquisition module; 22112. an image preprocessing module; 22113. GVF image generation module; 22114. an initial convergence position matching module; 22115. an iteration convergence module; 22116. an initial convergence profile position matching module; 22117. a contour point prediction module; 22118. a Snake iteration module; 22119. the identification result output module; 2212. an abnormality recognition module of the photovoltaic module; 22121. the photovoltaic module image acquisition module; 22122. an abnormal image acquisition module; 22123. an abnormal contour extraction module; 22124. an abnormal state identification module; 222. an operation state abnormality detection module; 23. a communication module; 24. a database; 3. a management terminal module; 31. an information inquiry module; 32. a status display module; 33. a control scheduling module; 34. and an abnormality alarm module.
Detailed Description
For the purpose of further illustrating the various embodiments, the present invention provides the accompanying drawings, which are a part of the disclosure of the present invention, and which are mainly used to illustrate the embodiments and, together with the description, serve to explain the principles of the embodiments, and with reference to these descriptions, one skilled in the art will recognize other possible implementations and advantages of the present invention, wherein elements are not drawn to scale, and like reference numerals are generally used to designate like elements.
According to an embodiment of the invention, a monitoring management system applied to photovoltaic power generation is provided.
The invention is further described with reference to the accompanying drawings and the specific embodiments, as shown in fig. 1 to fig. 4, a monitoring management system applied to photovoltaic power generation according to an embodiment of the invention comprises a monitoring terminal module 1, a cloud server 2 and a management terminal module 3; the monitoring terminal module 1 comprises a data acquisition module 11 and a video monitoring module 12;
the data acquisition module 11 is used for acquiring environmental data and operation data of the photovoltaic power station;
specifically, the data acquisition module 11 includes a temperature and humidity acquisition module 111, a current acquisition module 112, a voltage acquisition module 113 and a power acquisition module 114;
The temperature and humidity acquisition module 111 is used for acquiring real-time temperature and humidity of a photovoltaic module in a photovoltaic power station during operation of a field photovoltaic module; the current collection module 112 is used for collecting real-time operation current of the photovoltaic module and the photovoltaic grid-connected inverter; the voltage acquisition module 113 is used for acquiring real-time operation voltage of the photovoltaic module and the photovoltaic grid-connected inverter; the power collection module 114 is configured to collect real-time operating power of the photovoltaic module and the photovoltaic grid-connected inverter.
The video monitoring module 12 is used for carrying out real-time video monitoring on the field condition of the photovoltaic power station; the video monitoring in this embodiment can be implemented by using a monitoring camera, and the monitoring camera is installed right above the photovoltaic module through a vertical rod and is matched with a corresponding sliding track, so that the camera can move on the track, thereby implementing video monitoring on different photovoltaic modules, and the camera device can be installed on an unmanned aerial vehicle, thereby implementing video monitoring on the photovoltaic module through the flying unmanned aerial vehicle, or other monitoring methods.
The cloud server 2 comprises a data acquisition module 21, an analysis diagnosis module 22, a communication module 23 and a database 24;
the data acquisition module 21 is used for acquiring environmental data, operation data and on-site real-time video monitoring data of the photovoltaic power station;
The analysis and diagnosis module 22 is used for performing abnormality detection on the photovoltaic module of the photovoltaic power station by utilizing an image recognition technology, and is also used for performing analysis and detection on the operation state of the photovoltaic module by utilizing the acquired operation data;
specifically, the analysis and diagnosis module 22 includes a photovoltaic module abnormality detection module 221 and an operation state abnormality detection module 222;
the photovoltaic module abnormality detection module 221 is configured to perform abnormality detection on a photovoltaic module in a photovoltaic power station by using an image recognition technology, so as to implement abnormality detection on a state of the photovoltaic module;
the photovoltaic module abnormality detection module 221 includes a photovoltaic module identification module 2211 and a photovoltaic module abnormality identification module 2212;
the photovoltaic module identification module 2211 is used for identifying a photovoltaic module in field real-time video monitoring data by using an improved GVF snap model, and includes the following steps:
acquiring sequence images of a live real-time monitoring video, judging whether the sequence images have reflection phenomena, and performing image enhancement processing when the reflection phenomena occur;
adopting a PM model to carry out smoothing treatment on the sequence image after the enhancement treatment, and enhancing the edge to remove noise;
calculating by using the improved GVF model, and iteratively generating an improved image GVF;
Clicking a set key point position near the edge of the image in a mouse clicking manner along the edge of the photovoltaic module string in the first sequence image, and obtaining a convergence profile used for predicting a second layer of profile points and obtaining an initial convergence position through matching;
the contour obtained by matching is used as an initial contour of a snake, the position of a first sequence image is used as a predicted position of a second sequence image, and the position is also the matched initial position, the initial converged contour position is obtained after matching, whether the initial converged contour position exceeds a bottom layer is judged, if yes, identification of a photovoltaic module string in the sequence image is realized, and otherwise, the next step is executed;
predicting contour points of the next layer by utilizing convergence results of the first two layers, and performing block matching by taking the predicted positions as initial positions;
and carrying out iteration convergence by utilizing Snake with three-dimensional constraint, adding a layer number to carry out layer-crossing judgment again and again, and carrying out subsequent prediction and matching, thereby circulating until the identification of all sequence images is realized.
Specifically, the photovoltaic module recognition module 2211 includes a sequential image acquisition module 22111, an image preprocessing module 22112, a GVF image generation module 22113, an initial convergence position matching module 22114, an iteration convergence module 22115, an initial convergence contour position matching module 22116, a contour point prediction module 22117, a snap iteration module 22118, and a recognition result output module 22119;
The sequence image acquisition module 22111 is used for acquiring sequence images of the live real-time monitoring video;
the image preprocessing module 22112 is used for performing image enhancement processing when the sequence image has a reflection phenomenon, and is also used for performing smoothing processing on the sequence image by using a PM (Perona-Molik) model;
the GVF image generation module 22113 is configured to perform calculation using the improved GVF model, and iteratively generate an improved image GVF;
because the GVF model is easily misled by external forces of other objects, the adjacent parts of the two objects are mainly represented, and the GVF model is easily misled by external forces of another object during segmentation, so that the GVF model converges in opposite directions, and the segmentation effect cannot be achieved. Therefore, the GVF model is improved in this embodiment, and an improved GVF model is obtained, where the improvement is as follows:
the mathematical expression for the conventional GVF snap is:
this embodiment is modified as follows:
as is evident from the above equation, the smoothness constraint derived from the identity matrix does not take into account the edge structure information of the image, and finally an isotropic diffusion equation is obtained. It is naturally conceivable to replace W with a matrix D associated with the image edge domain, resulting in a new GVF model, i.e. an improved GVF model, which takes into account the edge structure information of the image and thus has an adaptive capacity expressed as:
Wherein u is x 、u y 、v x 、v y Is the first order bias derivative of u and v to x and y respectively, W is a unit matrixF represents the edge map of the processed image, (x, y) represents the coordinates of the pixels, and matrix D is +.>Is a symmetrical and semi-positive definite matrix,λ 1 and lambda (lambda) 2 The characteristic value is represented, mu represents a weighting coefficient, V represents a vector field, f represents a gradient field of f, u represents a projection of a force field in an x direction, V represents a projection of a force field in a y direction, V represents a gradient operator for image solving, dx represents a convolution kernel in the x direction, dy represents a convolution kernel in the y direction, T represents a transpose, and epsilon represents an energy function.
The initial convergence position matching module 22114 is configured to click a set key point position near an image edge in a mouse click manner along the edge of the photovoltaic module string in the first sequence image, and the obtained convergence profile is used to predict a second layer profile point and obtain an initial convergence position after matching;
the iteration convergence module 22115 is used for iterating by taking the contour of the initial convergence position as the initial contour of the Snake;
iterative convergence process:
1. the distance between the control points is suitably adjusted and if the distance is too small, the two control points are combined into one control point. If the interval between the adjacent control points is too large, the control points are added, so that the number of the control points can be kept within a reasonable range;
2. Calculating GVF (using the modified GVF model) from the current snake position as an external force;
3. calculating internal force E with three-dimensional constraint int ，E int Representing the internal energy of the movable contour line, wherein the smooth continuity of the contour is controlled by the internal energy;
4. calculation of the resultant force E snake =E int + E image +E constrain Wherein E is image Representing the degree of coincidence between the active contour and the image feature, e.g. the gray level or gradient of the point, the main effect being to push the contour to slide on the image plane, E constrain From prior information or information provided by a user, the contour line can be forced to necessarily pass through some points with specific significance on the image;
5. moving the control point according to the resultant force calculated in the step 4;
6. and judging the convergence condition.
The initial convergence contour position matching module 22116 is configured to take the first sequence image position as a predicted position of the second sequence image, and also as a matched initial position, obtain an initial convergence contour position after matching, and determine whether the initial convergence contour position exceeds a bottom layer, if so, identify a photovoltaic module string in the sequence image, and if not, perform block matching;
specifically, in order to correctly estimate the position and shape of the object contour in the current image, information of the previous image may be used, that is, the position, shape, and their trend of change of the object contour in the previous image may be used. The initial position is set by adopting a method of combining prediction and matching. In this embodiment, consistency of position change of the control point is used for representing consistency and smoothness of object change during segmentation.
Based on the prediction, a matching algorithm is applied in this embodiment to increase the accuracy of the initial contour. According to the characteristic that video images and three-dimensional image sequences have a series of similarity, motion estimation algorithms based on block matching widely used in H.26X and MPEG-X are applied to the segmentation of three-dimensional images. According to a certain matching criterion, the method searches the block with the greatest similarity with the current block in different frames according to a certain searching method, namely the matching block.
The contour point prediction module 22117 is configured to predict a contour point of a next layer by using a convergence result of the first two layers, and perform block matching with the predicted position as an initial position;
the Snake iteration module 22118 is configured to perform iteration convergence by using Snake with three-dimensional constraint, perform layer-by-layer judgment again and again, and perform subsequent prediction and matching, so as to circulate until recognition of all sequence images is achieved;
the identification result output module 22119 is configured to output identification results of the photovoltaic module strings in all the sequence images, and number all the photovoltaic module strings.
The image preprocessing module 22112 comprises a reflection judgment module, an image enhancement processing module and an image smoothing processing module;
The reflection judging module is used for judging whether the sequence image has a reflection phenomenon or not; specifically, the judgment concept of the reflection judgment is as follows: and traversing the whole image by adopting image blocks, wherein the image blocks adopt squares with side lengths of 25 pixel points, judging all pixels in each image block in the traversing process, if the white pixel points in the image blocks occupy more than half of the total pixel points of the image blocks, considering the image blocks as reflective image blocks, judging the number of the reflective image blocks after the whole image is traversed, and if the number of the reflective image blocks exceeds 1% of the total number of the image blocks, judging that the image has partial reflection.
The image enhancement processing module is used for carrying out image enhancement processing when the light reflection phenomenon occurs in the column images; specifically, the image enhancement processing is realized by adopting a contrast adjustment algorithm, wherein the adjustment formula of the contrast adjustment algorithm is as follows:
y=(x-127.5)*tan[(45+44*c)/180*PI]+127.5
wherein y represents an output pixel, x represents an input pixel, PI represents a circumference ratio, the value range of the parameter c is [ -1,1], and tan represents a tangent function;
the image smoothing processing module is used for smoothing the sequence images, enhancing edges and removing noise.
The photovoltaic module abnormality recognition module 2212 is used for recognizing and classifying foreign matter shielding or module breakage conditions of the photovoltaic module by utilizing a probability Hough transform algorithm and a contour extraction algorithm.
Specifically, the photovoltaic module abnormality recognition module 2212 includes a photovoltaic module image acquisition module 22121, an abnormality image acquisition module 22122, an abnormality contour extraction module 22123, and an abnormality state recognition module 22124;
the photovoltaic module image acquisition module 22121 is used for intercepting a photovoltaic module area in the minimum circumscribed rectangle of the module according to the identification result of the photovoltaic module to obtain a single photovoltaic module image;
the abnormal image acquisition module 22122 is used for eliminating a background plate and grid lines of a single photovoltaic module image based on morphology, and eliminating a frame of the single photovoltaic module image by using a probability Hough transform algorithm to obtain an abnormal image;
specifically, the elimination of the background plate and the grid lines includes: firstly, converting a photovoltaic module image into an HSV image format, extracting a blue background plate according to an HSV color model, then removing the blue background plate from the image to obtain a binary image, and finally eliminating grid line interference remained in the binary image by using morphological closing operation.
The elimination of the frame comprises: firstly, carrying out edge detection on a binary image for eliminating a background plate and grid lines by adopting a Canny algorithm, extracting edge information at the position of a frame, then carrying out frame straight line detection by utilizing probability Hough transformation, and finally, eliminating the frame.
The abnormal contour extraction module 22123 is used for extracting the contour of the abnormal image by utilizing an eight-domain traversal algorithm to obtain an extracted contour map;
the abnormal state identification module 22124 is configured to identify and classify an abnormal state of the photovoltaic module based on the extracted contour map, and specifically includes:
judging whether the effective contour is extracted, if not, indicating that no abnormality exists, if yes, judging whether a plurality of contours exist, if not, judging whether to execute color features in the contours, and if yes, executing the next step;
after the screening threshold is reduced, continuously judging whether radial areas appear on the multiple outlines, if so, judging that the photovoltaic module glass is damaged, and if not, judging that the multiple foreign matters are shielded;
judging whether the color features in the outline are non-white areas, if so, judging that the outline is shielded by the foreign matter, if not, judging that the outline has regularity, if so, judging that the EVA falls off, and if not, judging that the outline is shielded by the foreign matter.
When it is determined that the foreign matter is blocked, the method comprises the following steps:
extracting the minimum circumscribed rectangle of the extracted contour image to obtain the minimum circumscribed rectangle of all contours, and simultaneously calculating the length, width, length-width ratio and area information of all the minimum circumscribed rectangle;
Screening and removing all minimum circumscribed rectangles based on a preset contour screening condition, screening out areas belonging to foreign matter shielding, and obtaining the contour of the foreign matter shielding and the position information of the shielding areas;
because some of the outline is the smallest circumscribed rectangle of the interference outline, outline screening conditions are formulated according to the length, width, area and other differences among different circumscribed rectangles, all the smallest circumscribed rectangles are screened and excluded, the area which belongs to shielding is screened out, and the outline screening conditions are as follows:
condition 1: width is greater than or equal to LTH & Length is less than or equal to HTH & Length is greater than or equal to Width
Condition 2: a is less than or equal to length/Width is less than or equal to b
In the formula, width represents the minimum circumscribed rectangle Width, length represents the minimum circumscribed rectangle length, LTH represents the low threshold value size, HTH represents the high threshold value size, a and b represent range parameters, and when the minimum circumscribed rectangle meets the condition 1 and the condition 2 at the same time, the minimum circumscribed rectangle is considered to be a foreign matter shielding area.
The operation state anomaly detection module 222 is configured to analyze and detect an operation state of the photovoltaic module by using the improved BP wavelet neural network model in combination with the acquired operation data, so as to implement anomaly detection of the operation state of the photovoltaic module, and specifically includes:
Constructing an improved BP wavelet neural network model, and training the model by utilizing historical operation data of a photovoltaic power station in a database;
collecting operation data of the photovoltaic module and the photovoltaic grid-connected inverter at the previous moment, and inputting the operation data into a trained improved BP wavelet neural network model to obtain predicted operation data of the photovoltaic module and the photovoltaic grid-connected inverter at the current moment;
and analyzing and comparing the operation data of the photovoltaic module and the photovoltaic grid-connected inverter obtained by detection at the current moment with the predicted operation data, and judging that the operation state of the photovoltaic module at the current moment is abnormal when the difference value between the operation data and the predicted operation data exceeds a preset threshold value.
Specifically, the construction of the improved BP wavelet neural network model comprises the following steps:
setting an input layer to a hidden layer weight value, a hidden layer neuron threshold value and a wavelet expansion translation parameter by combining wavelet type, time-frequency parameters and training sample data;
constructing a new error function by introducing hidden layer saturation, and establishing an improved BP wavelet neural network model;
wherein the new error function is formulated as follows:
wherein P represents the total number of input samples, P represents the samples, m represents the number of nodes of the output layer, j represents each node of the output layer, l represents the number of nodes of the hidden layer, k represents each node of the hidden layer, N represents an even number, and t p Indicating the desired output, t pj Representing the expected output of each node of the output layer, t pk Representing the expected output of each node of the hidden layer, y p The actual output is indicated as such,y pj representing the actual output of each node of the output layer,representing the hidden layer saturation scale factor, E A Represents the A-th sample error, E B Indicating sample B error.
Inputting layer to hidden layer weightsThe expression of (2) is as follows:
in the method, in the process of the invention,representing input layer to implicit layer weight +.>Normalized one by one and multiplied by the weight of the hidden layer node number l, the input layer node number n and the transfer function related factors, x imax Representing the maximum value of the samples in the ith neuron of the input layer, x imin Representing a sample minimum in an input layer i-th neuron;
hidden layer neuron thresholdThe expression of (2) is as follows:
in the method, in the process of the invention,representing hidden layer neuron threshold->A threshold value multiplied by a factor associated with the hidden layer node number l, the input layer node number n, and the transfer function;
wavelet expansion translation parameter a k And b k The expression of (2) is as follows:
wherein t is * Representing the time domain center of the wavelet,representing the radius.
The communication module 23 is used for realizing wireless communication connection between the cloud server and the monitoring terminal module and between the cloud server and the management terminal module;
the database 24 is used for storing and calling data generated in the running process of the system;
The management terminal module 3 comprises an information inquiry module 31, a state display module 32, a control scheduling module 33 and an abnormality alarm module 34;
the information inquiry module 31 is used for an administrator to inquire the environment data, the operation data and the on-site real-time video monitoring data of the photovoltaic power station;
the status display module 32 is used for displaying environmental data, operation data and on-site real-time video monitoring data of the photovoltaic power station;
the control scheduling module 33 is configured to execute a corresponding control scheduling plan according to the collected operation data and the detected abnormal data;
the abnormality alarm module 34 is used for alarming when the operation state of the photovoltaic power plant is abnormal, and sending a maintenance request notification to a manager.
The abnormality alarming module 34 comprises a voice alarming module and a maintenance requesting module, wherein the voice alarming module is used for giving a voice alarm when detecting that the photovoltaic module of the photovoltaic power station is abnormal; the maintenance request module is used for sending a request for maintenance to the manager after the abnormality is detected.
In summary, by means of the technical scheme, through the cooperation among the monitoring terminal module, the cloud server and the management terminal module, the collection, storage and analysis processing of the operation parameters and the real-time state data of the photovoltaic module can be realized, so that the operation parameter prediction, fault display, alarm prompt and the like of the photovoltaic power generation system can be realized, and further, the comprehensive monitoring and management functions can be provided for users.
In addition, through being provided with photovoltaic module anomaly detection module and the unusual detection module of running state to can not only utilize image recognition technology to realize the detection and the discernment of phenomenon such as photovoltaic module surface foreign matter shelters from, the damaged or material drop of subassembly under the unusual detection module's of photovoltaic module effect, can also utilize the unusual detection module of running state to utilize modified BP wavelet neural network model to realize the detection and the discernment to unusual operation data, and then can improve photovoltaic power generation system's operation safety effectively.
In addition, through being provided with photovoltaic module identification module and photovoltaic module unusual identification module to can utilize the photovoltaic module in the real-time video monitoring data of modified GVF Snake model to discern the photovoltaic module in scene under photovoltaic module identification module's effect, can be better at the image segmentation consider the marginal information of image, make the separation effect more ideal, can also utilize probability Hough transform algorithm to combine the profile extraction algorithm to realize the discernment of phenomenon such as photovoltaic module surface foreign matter shelters from, subassembly damage or material drop under photovoltaic module unusual identification module's effect, and then can be better satisfy the control management demand of photovoltaic power plant.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (6)
1. The monitoring management system for the photovoltaic power generation is characterized by comprising a monitoring terminal module (1), a cloud server (2) and a management terminal module (3);
the monitoring terminal module (1) comprises a data acquisition module (11) and a video monitoring module (12);
the cloud server (2) comprises a data acquisition module (21), an analysis and diagnosis module (22), a communication module (23) and a database (24);
the data acquisition module (21) is used for acquiring environmental data, operation data and on-site real-time video monitoring data of the photovoltaic power station;
the analysis and diagnosis module (22) is used for carrying out abnormality detection on the photovoltaic module of the photovoltaic power station by utilizing an image recognition technology and also is used for carrying out analysis and detection on the operation state of the photovoltaic module by utilizing the acquired operation data;
the management terminal module (3) comprises an information inquiry module (31), a state display module (32), a control scheduling module (33) and an abnormality alarm module (34);
The control scheduling module (33) is used for executing a corresponding control scheduling plan according to the collected operation data and the detected abnormal data;
the abnormality alarm module (34) is used for alarming when the operation state of the photovoltaic power station is abnormal and sending a maintenance request notice to a manager;
the analysis and diagnosis module (22) comprises a photovoltaic module abnormality detection module (221) and an operation state abnormality detection module (222);
the photovoltaic module abnormality detection module (221) is used for detecting abnormality of a photovoltaic module in the photovoltaic power station by utilizing an image recognition technology, so as to realize abnormality detection of the state of the photovoltaic module;
the operation state abnormality detection module (222) is used for analyzing and detecting the operation state of the photovoltaic module by utilizing the improved BP wavelet neural network model and combining the acquired operation data so as to realize abnormality detection of the operation state of the photovoltaic module;
the photovoltaic module abnormality detection module (221) comprises a photovoltaic module identification module (2211) and a photovoltaic module abnormality identification module (2212);
the photovoltaic module identification module (2211) is used for identifying the photovoltaic module in the field real-time video monitoring data by utilizing the improved GVF Snake model;
The photovoltaic module abnormality identification module (2212) is used for identifying and classifying foreign matter shielding or module breakage conditions of the photovoltaic module by utilizing a probability Hough transform algorithm and a contour extraction algorithm;
the photovoltaic module identification module (2211) comprises a sequence image acquisition module (22111), an image preprocessing module (22112), a GVF image generation module (22113), an initial convergence position matching module (22114), an iteration convergence module (22115), an initial convergence contour position matching module (22116), a contour point prediction module (22117), a Snake iteration module (22118) and an identification result output module (22119);
the sequence image acquisition module (22111) is used for acquiring sequence images of the live real-time monitoring video;
the image preprocessing module (22112) is used for carrying out image enhancement processing when the sequence image has a reflection phenomenon and also used for carrying out smoothing processing on the sequence image;
the GVF image generation module (22113) is used for carrying out calculation by utilizing the improved GVF model and iteratively generating an improved GVF image;
the expression of the improved GVF model is:
where ε represents the energy function, f represents the edge map of the processed image, (x, y) represents the coordinates of the pixels, D represents a symmetric and semi-positive matrix, μ represents the weighting coefficient, V represents the vector field, f represents the gradient field of f, u represents the projection of the force field in the x-direction, V represents the projection of the force field in the y-direction, V represents the gradient operator of the image solution, dx represents the convolution kernel in the x-direction, dy represents the convolution kernel in the y-direction, and T represents the transpose;
The initial convergence position matching module (22114) is used for clicking a set key point position near the edge of the image in a mouse clicking manner along the edge of the photovoltaic module string in the first sequence image, and the obtained convergence profile is used for predicting a second layer profile point and obtaining an initial convergence position through matching;
the iteration convergence module (22115) is used for iterating by taking the contour of the initial convergence position as the initial contour of the Snake;
the initial convergence contour position matching module (22116) is used for taking the position of the first sequence image as the predicted position of the second sequence image, obtaining the initial convergence contour position after matching, judging whether the initial convergence contour position exceeds the bottom layer, if so, realizing the identification of the photovoltaic module string in the sequence image, and if not, performing block matching;
the contour point prediction module (22117) is used for predicting a contour point of a next layer by utilizing convergence results of the first two layers and performing block matching by taking the predicted position as an initial position;
the Snake iteration module (22118) is used for carrying out iteration convergence by utilizing Snake with three-dimensional constraint, carrying out layer-by-layer judgment again and again, and carrying out subsequent prediction and matching, so as to circulate until the identification of all sequence images is realized;
The identification result output module (22119) is used for outputting identification results of the photovoltaic module strings in all the sequence images and numbering all the photovoltaic module strings.
2. The monitoring and management system for photovoltaic power generation according to claim 1, wherein the data acquisition module (11) comprises a temperature and humidity acquisition module (111), a current acquisition module (112), a voltage acquisition module (113) and a power acquisition module (114).
3. The monitoring and management system for photovoltaic power generation according to claim 1, wherein the image preprocessing module (22112) comprises a light reflection judging module, an image enhancement processing module and an image smoothing processing module;
the reflection judging module is used for judging whether the sequence image has a reflection phenomenon or not;
the image enhancement processing module is used for carrying out image enhancement processing when the light reflection phenomenon occurs in the column images;
the image smoothing processing module is used for carrying out smoothing processing on the sequence images, enhancing edges and removing noise.
4. The monitoring and management system for photovoltaic power generation according to claim 3, wherein the photovoltaic module abnormality identification module (2212) comprises a photovoltaic module image acquisition module (22121), an abnormality image acquisition module (22122), an abnormality profile extraction module (22123) and an abnormality state identification module (22124);
The photovoltaic module image acquisition module (22121) is used for intercepting a photovoltaic module area in the minimum circumscribed rectangle of the module according to the identification result of the photovoltaic module to obtain a single photovoltaic module image;
the abnormal image acquisition module (22122) is used for eliminating a background plate and grid lines of a single photovoltaic module image based on morphology, and eliminating a frame of the single photovoltaic module image by using a probability Hough transform algorithm to obtain an abnormal image;
the abnormal contour extraction module (22123) is used for carrying out contour extraction on the abnormal image by utilizing an eight-domain traversing algorithm to obtain an extracted contour map;
the abnormal state identification module (22124) is used for identifying and classifying the abnormal state of the photovoltaic module based on the extracted profile;
wherein, the abnormal state identification module (22124) includes when identifying and classifying the abnormal state of the photovoltaic module based on the extracted profile map:
judging whether the effective contour is extracted, if not, indicating that no abnormality exists, if yes, judging whether a plurality of contours exist, if not, judging whether to execute color features in the contours, and if yes, executing the next step;
after the screening threshold is reduced, continuously judging whether radial areas appear on the multiple outlines, if so, judging that the photovoltaic module glass is damaged, and if not, judging that the multiple foreign matters are shielded;
Judging whether the color features in the outline are non-white areas, if so, judging that the outline is shielded by the foreign matter, if not, judging that the outline has regularity, if so, judging that the EVA falls off, and if not, judging that the outline is shielded by the foreign matter.
5. The monitoring and management system for photovoltaic power generation according to claim 4, wherein when it is determined that foreign matter is blocked, comprising the steps of:
extracting the minimum circumscribed rectangle of the extracted contour image to obtain the minimum circumscribed rectangle of all contours, and simultaneously calculating the length, width, length-width ratio and area information of all the minimum circumscribed rectangle;
screening and removing all minimum circumscribed rectangles based on a preset contour screening condition, screening out areas belonging to foreign matter shielding, and obtaining the contour of the foreign matter shielding and the position information of the shielding areas;
wherein, the profile screening conditions are:
condition 1: width is greater than or equal to LTH & Length is less than or equal to HTH & Length is greater than or equal to Width
Condition 2: a is less than or equal to length/Width is less than or equal to b
In the formula, width represents the minimum circumscribed rectangle Width, length represents the minimum circumscribed rectangle length, LTH represents the low threshold value size, HTH represents the high threshold value size, a and b represent range parameters, and when the minimum circumscribed rectangle meets the condition 1 and the condition 2 at the same time, the minimum circumscribed rectangle is considered to be a foreign matter shielding area.
6. The monitoring and management system for photovoltaic power generation according to claim 1, wherein the operation state anomaly detection module (222) performs analysis and detection on the operation state of the photovoltaic module by using the improved BP wavelet neural network model in combination with the acquired operation data, and implementing anomaly detection on the operation state of the photovoltaic module includes:
constructing an improved BP wavelet neural network model, and training the model by utilizing historical operation data of a photovoltaic power station in a database;
collecting operation data of the photovoltaic module and the photovoltaic grid-connected inverter at the previous moment, and inputting the operation data into a trained improved BP wavelet neural network model to obtain predicted operation data of the photovoltaic module and the photovoltaic grid-connected inverter at the current moment;
analyzing and comparing the operation data of the photovoltaic module and the photovoltaic grid-connected inverter obtained by detection at the current moment with the predicted operation data, and judging that the operation state of the photovoltaic module at the current moment is abnormal when the difference value between the operation data and the predicted operation data exceeds a preset threshold value;
the method for constructing the improved BP wavelet neural network model comprises the following steps of:
setting an input layer to a hidden layer weight value, a hidden layer neuron threshold value and a wavelet expansion translation parameter by combining wavelet type, time-frequency parameters and training sample data;
Constructing a new error function by introducing hidden layer saturation, and establishing an improved BP wavelet neural network model;
wherein the new error function is formulated as follows:
wherein P represents the total number of input samples, P represents the samples, m represents the number of nodes of the output layer, j represents each node of the output layer, l represents the number of nodes of the hidden layer, k represents each node of the hidden layer, N represents an even number, and t p Indicating the desired output, t pj Representing the expected output of each node of the output layer, t pk Representing the expected output of each node of the hidden layer, y p Representing the actual output, y pj Representing the actual output of each node of the output layer, +.>Representing the hidden layer saturation scale factor, E A Represents the A-th sample error, E B Indicating sample B error. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310455131.8A CN116191680B (en) | 2023-04-25 | 2023-04-25 | Monitoring management system applied to photovoltaic power generation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310455131.8A CN116191680B (en) | 2023-04-25 | 2023-04-25 | Monitoring management system applied to photovoltaic power generation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116191680A CN116191680A (en) | 2023-05-30 |
| CN116191680B true CN116191680B (en) | 2023-07-18 |
Family
ID=86452545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310455131.8A Active CN116191680B (en) | 2023-04-25 | 2023-04-25 | Monitoring management system applied to photovoltaic power generation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116191680B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117691592A (en) * | 2023-12-14 | 2024-03-12 | 湖南防灾科技有限公司 | Photovoltaic output prediction method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104680498A (en) * | 2015-03-24 | 2015-06-03 | 江南大学 | Medical image segmentation method based on improved gradient vector flow model |
| CN107358259A (en) * | 2017-07-13 | 2017-11-17 | 国家电网公司 | Covering ice for insulator detection method based on GLOH descriptions and GVF Snake models |
| CN113989261A (en) * | 2021-11-09 | 2022-01-28 | 浙江浙能嘉华发电有限公司 | Unmanned aerial vehicle visual angle infrared image photovoltaic panel boundary segmentation method based on Unet improvement |
| CN114825636A (en) * | 2022-05-26 | 2022-07-29 | 深圳博浩远科技有限公司 | Health state monitoring and warning system and method for photovoltaic inverter |
-
2023
- 2023-04-25 CN CN202310455131.8A patent/CN116191680B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104680498A (en) * | 2015-03-24 | 2015-06-03 | 江南大学 | Medical image segmentation method based on improved gradient vector flow model |
| CN107358259A (en) * | 2017-07-13 | 2017-11-17 | 国家电网公司 | Covering ice for insulator detection method based on GLOH descriptions and GVF Snake models |
| CN113989261A (en) * | 2021-11-09 | 2022-01-28 | 浙江浙能嘉华发电有限公司 | Unmanned aerial vehicle visual angle infrared image photovoltaic panel boundary segmentation method based on Unet improvement |
| CN114825636A (en) * | 2022-05-26 | 2022-07-29 | 深圳博浩远科技有限公司 | Health state monitoring and warning system and method for photovoltaic inverter |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116191680A (en) | 2023-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019214268A1 (en) | Photovoltaic array fault diagnosis method based on composite information | |
| CN112734692A (en) | Transformer equipment defect identification method and device | |
| CN112233073A (en) | Real-time detection method for infrared thermal imaging abnormity of power transformation equipment | |
| CN111754498A (en) | Conveyor belt carrier roller detection method based on YOLOv3 | |
| CN109359697A (en) | Graph image recognition methods and inspection system used in a kind of power equipment inspection | |
| CN111080691A (en) | Infrared hot spot detection method and device for photovoltaic module | |
| CN116191680B (en) | Monitoring management system applied to photovoltaic power generation | |
| CN114743115A (en) | Shelter recognition method and fault early warning system for photovoltaic electronic equipment | |
| Prabhakaran et al. | Deep Learning-Based Model for Defect Detection and Localization on Photovoltaic Panels. | |
| CN116690613B (en) | Control method and system of photovoltaic intelligent cleaning robot | |
| CN115642877A (en) | Photovoltaic module abnormal shielding detection method and system based on deep learning | |
| Zhang et al. | Aerial image analysis based on improved adaptive clustering for photovoltaic module inspection | |
| Hao et al. | Detection of bird nests on power line patrol using single shot detector | |
| CN116228670A (en) | Yolov 5-based photovoltaic defect detection method, device, computer equipment and medium | |
| Aakroum et al. | Deep learning for inferring the surface solar irradiance from sky imagery | |
| CN116046796A (en) | Photovoltaic module hot spot detection method and system based on unmanned aerial vehicle | |
| Otamendi et al. | A scalable framework for annotating photovoltaic cell defects in electroluminescence images | |
| Xia et al. | A multi-target detection based framework for defect analysis of electrical equipment | |
| Gu et al. | Object detection of overhead transmission lines based on improved YOLOv5s | |
| CN116454882B (en) | Photovoltaic power generation prediction method based on machine vision predictor | |
| Hou et al. | Multi angle location and identification method of suspension insulators based on R2CNN algorithm | |
| Balaji et al. | A Framework for Fault Recognition of Photovoltaic Plants with Convolutional Neural Network | |
| CN117237301A (en) | Redundant information decoupling-based power grid defect equipment identification algorithm | |
| Wu et al. | Research on State Recognition of Platen Based on Improved K-means Algorithm | |
| CN108470140B (en) | Power transmission line bird nest identification method based on statistical characteristics and machine learning |
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 |