CN114494830B - Multi-source information photovoltaic map generation method and device - Google Patents
Multi-source information photovoltaic map generation method and device Download PDFInfo
- Publication number
- CN114494830B CN114494830B CN202210063493.8A CN202210063493A CN114494830B CN 114494830 B CN114494830 B CN 114494830B CN 202210063493 A CN202210063493 A CN 202210063493A CN 114494830 B CN114494830 B CN 114494830B
- Authority
- CN
- China
- Prior art keywords
- photovoltaic
- data
- model
- pvi
- map
- 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
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 42
- 238000007726 management method Methods 0.000 claims abstract description 32
- 238000012795 verification Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 32
- 238000013527 convolutional neural network Methods 0.000 claims description 14
- 238000001514 detection method Methods 0.000 claims description 13
- 238000011161 development Methods 0.000 claims description 8
- 239000003086 colorant Substances 0.000 claims description 6
- 238000002790 cross-validation Methods 0.000 claims description 6
- 238000005520 cutting process Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000003032 molecular docking Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 8
- 238000010248 power generation Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Classifications
-
- 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
-
- 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
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of remote sensing, and particularly provides a multi-source information photovoltaic map generation method, which utilizes correlation among multiple bands of remote sensing data, combines actual operation data of a photovoltaic energy management platform, designs a photovoltaic facility identification model and a photovoltaic facility paving recommendation model based on a remote sensing image top view, acquires photovoltaic paving query analysis results by querying photovoltaic paving related data from internet images and utilizing an image text photovoltaic analysis identification model; and (3) performing target identification and mutual cross verification by integrating multiparty data, and constructing a photovoltaic map based on the remote sensing image map, the additional real-time operation data and the photovoltaic paving recommendation to form more accurate photovoltaic system operation and planning. Compared with the prior art, the photovoltaic map is constructed by using the photovoltaic energy management platform, so that finer photovoltaic system operation is formed, and the platform operation efficiency is improved.
Description
Technical Field
The invention relates to the technical field of remote sensing, and particularly provides a method and a device for generating a multi-source information photovoltaic map.
Background
With the rapid development of deep learning technology and the support of mass data and high-efficiency computing capacity in the Internet and cloud computing age, the deep learning technology represented by CNN convolutional neural networks is trained and constructed to obtain a large-scale neural network similar to a human brain structure, and breakthrough progress is made in the fields of computer vision, voice recognition, natural language understanding and the like, and subversion change is brought to the whole society.
In recent years, the remote sensing technology is widely applied, multispectral images and full-color images obtained by satellite shooting are formed into remote sensing images with higher spatial resolution and spectral resolution through image fusion, and the remote sensing technology has advantages in acquiring basic geographic data, resource information and emergency disaster data compared with other technical means, and is widely applied to the fields of national economy and military.
The problems of environmental pollution, climate change and the like caused by the large-scale development and utilization of traditional fossil energy are increasingly prominent in the global scope, and solar energy is the cleanest, safe and reliable energy in the future, and the development and utilization of the solar energy are planned for a long time as the main content of the energy revolution. Photovoltaic (photovoltaics) is a short term for solar Photovoltaic power generation systems (Solar power system), a novel power generation system that uses the Photovoltaic effect of solar cell semiconductor materials to directly convert solar radiant energy into electrical energy. The photovoltaic has the characteristics of relatively small distribution, relatively small scale, uncertainty of power output, randomness and the like, so that the controllability is obviously reduced, the photovoltaic is required to be subjected to finer management and layout, meanwhile, other factors such as the safety of the photovoltaic and light pollution are comprehensively considered, higher requirements are put forward on operation management of the photovoltaic, a photovoltaic map is built by means of a remote sensing technology to realize reasonable planning and accurate management of the photovoltaic to become research hotspots, and the unified and accurate management of photovoltaic facilities based on the position and the map is realized while guiding services are provided for development contractors by displaying information such as sunshine and shadow states of a roof, photovoltaic resources, development mode selection and the like.
Under the condition, how to effectively integrate big data, remote sensing technology, deep learning technology and other technologies to generate a photovoltaic map, and providing support for reasonable planning and accurate management of the photovoltaic becomes a problem to be solved urgently.
Disclosure of Invention
The invention aims at the defects of the prior art and provides a multi-source information photovoltaic map generation method with strong practicability
The invention further aims to provide a multi-source information photovoltaic map generation device which is reasonable in design, safe and applicable.
The technical scheme adopted for solving the technical problems is as follows:
According to the method, correlation among multiple bands of remote sensing data is utilized, actual operation data of a photovoltaic energy management platform is combined, a photovoltaic facility identification model and a photovoltaic facility paving recommendation model are designed based on a remote sensing image top view, photovoltaic paving related data from an Internet image are inquired, and a photovoltaic paving inquiry analysis result in the photovoltaic facility identification model is obtained by utilizing an image character photovoltaic analysis identification model;
and (3) performing target identification and mutual cross verification by integrating multiparty data, and constructing a photovoltaic map based on the remote sensing image map, the additional real-time operation data and the photovoltaic paving recommendation to form more accurate photovoltaic system operation and planning.
Further, the remote sensing image is an image picture generated by image fusion of a multispectral image and a full-color image obtained through satellite shooting; the photovoltaic energy management platform is deployed in a cloud data center for providing photovoltaic facility data acquisition, and the cloud data center meets the resource requirements required by the operation of the photovoltaic energy management platform;
the core of the photovoltaic facility identification model RS-PVI-Det is a target detection model of a CNN convolutional neural network, and a photovoltaic facility target area is identified by inputting a remote sensing image for target detection;
The core of the photovoltaic facility paving recommendation model RS-PVI-Pred is a target detection model based on a CNN convolutional neural network, and based on remote sensing images, the region suitable for photovoltaic paving and the degree suitable for paving the photovoltaic facilities are output;
The IMG-PVI-Det core is a network model based on text recognition and image recognition and performing comprehensive analysis, semantic data in the network model are extracted, and comprehensive analysis and recognition are performed by combining the image recognition result to obtain a photovoltaic laying inquiry analysis result;
cross-verifying to obtain a Cross-verifying model, wherein the Cross-Check is based on output results from the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det based on the same coordinate position area, and a judgment result formed by judging the Cross-verifying model by adopting a mode of combining a preset rule with a neural network;
The photovoltaic Map generation module PVI-Map-Gen is used for constructing a photovoltaic Map by adding operation data such as real-time acquisition data and power generation capacity of the photovoltaic facilities on the basis of the identification result and the data of actual photovoltaic facilities managed by the photovoltaic energy management platform on the basis of the remote sensing image Map, providing recommended suggestion of a photovoltaic paving area and displaying planning prediction data such as expected income, power generation capacity and the like.
Further, training for the photovoltaic facility identification model includes:
S101, collecting remote sensing image data and marking photovoltaic facilities;
S102, designing a photovoltaic facility identification model RS-PVI-Det based on a target detection model;
S103, performing image cutting on remote sensing image data according to the input size of the photovoltaic facility identification model RS-PVI-Det;
s104, training the photovoltaic facility identification model RS-PVI-Det to form the photovoltaic facility identification model RS-PVI-Det.
Further, training of the photovoltaic facility paving recommendation model includes:
Training of a photovoltaic facility lay recommendation model, comprising:
s201, collecting related data of photovoltaic facilities of actual planning operation provided by a photovoltaic energy management platform;
S202, marking remote sensing images before paving the actually operated photovoltaic facilities provided by the photovoltaic energy management platform based on actual paving conditions;
S203, marking data of a photovoltaic facility area which is provided by the photovoltaic energy management platform and is planned and built on a remote sensing image;
s204, designing a photovoltaic facility paving recommendation model RS-PVI-Pred;
s205, performing image cutting on remote sensing image data according to the input size of a RS-PVI-Pred of a photovoltaic facility paving recommendation model;
S206, training the photovoltaic facility paving recommendation model RS-PVI-Pred to form the photovoltaic facility paving recommendation model RS-PVI-Pred.
Further, training of the image text photovoltaic analysis recognition model comprises the following steps:
S301, collecting related data with coordinate position data on a digital medium through an internet photovoltaic information inquiry module IE-PVI-Search;
s302, cleaning and filtering grouping is carried out on the collected data, and data marking is carried out;
S303, designing an image character photovoltaic analysis and identification model IMG-PVI-Det;
S304, carrying out joint training on the model by taking the marked keyword data as input to form an image character photovoltaic analysis recognition model IMG-PVI-Det.
Further, the cross-validation model training includes:
S401, forming a basic input data set by the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det based on the output result of the same coordinate position area based on remote sensing image data and internet digital media data;
S402, performing cross-validation result judgment data annotation on a basic input data set to form a data set;
s403, designing the Cross-Check model;
S404, fixing network parameters of the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det, and performing joint training on a Cross-Check model;
S405, extracting set business rules according to actual photovoltaic facility data of the photovoltaic energy platform, and combining the Cross-Check model of the combined training to form the Cross-Check model of the combined training.
Further, the step of generating the photovoltaic map includes:
S501, setting a map generation area position range;
s502, identifying a photovoltaic facility target area through the photovoltaic facility identification model RS-PVI-Det based on remote sensing image data and a set area position range;
S503, searching digital media in a set area through an internet photovoltaic information inquiry module IE-PVI-Search in the remote sensing image data and the set area range, and analyzing and identifying based on the image text photovoltaic analysis and identification model IMG-PVI-Det to obtain a target result;
S504, judging the result output by the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det by combining the actual photovoltaic facility data of the photovoltaic energy platform and the data in the same area through the Cross-verification model Cross-Check;
s505, outputting a recommended photovoltaic facility paving region and a recommended degree value by using the photovoltaic facility paving recommended model RS-PVI-Pred based on remote sensing image data and a set region position range;
s506, according to the output from the step S502 to the step S505, generating a photovoltaic map based on the remote sensing image data and the position data and combining the actual photovoltaic facility data and the planning data of the photovoltaic energy platform.
Further, the application and optimization of the photovoltaic map comprises:
s601, acquiring data of photovoltaic facilities in real time by using a photovoltaic energy platform, and attaching the data to a map for display and continuously updating;
S602, collecting data of photovoltaic facilities in real time by a photovoltaic energy platform, adding the data to a map for display, distinguishing the data by colors, and providing related estimation data for decision planning;
s603, docking an external photovoltaic energy platform, accessing data to the platform, attaching the data to a photovoltaic map according to the position data for display, and distinguishing the data by colors;
S604, according to different attention points, personalized data display is carried out, information such as roof sunshine and shadow states, photovoltaic resources and development mode selection is displayed, actual feedback is collected, a model is continuously optimized, and a more reasonable and accurate photovoltaic map is generated.
A multi-source information photovoltaic map generation apparatus, comprising: at least one memory and at least one processor;
the at least one memory for storing a machine readable program;
The at least one processor is configured to invoke the machine-readable program to perform a multi-source information photovoltaic map generation method.
Compared with the prior art, the method and the device for generating the multi-source information photovoltaic map have the following outstanding beneficial effects:
According to the invention, the spectral characteristics of the remote sensing image and the conventional selection position of photovoltaic laying are fully considered, a photovoltaic facility identification model and a photovoltaic facility laying recommendation model are designed based on a remote sensing image top view and combined with a deep learning target detection algorithm, and related data of digital media such as microblogs, blogs, news and the like from the Internet on the photovoltaic, roof, top view, unmanned aerial vehicle and the like are combined, meanwhile, operation planning data of a photovoltaic energy management platform are considered, so that a more accurate photovoltaic map is formed, reasonable planning and accurate management of the photovoltaic are realized, photovoltaic illegal construction engineering is actively found, the laying safety of the photovoltaic facilities is ensured, and the light pollution is reduced.
Compared with the traditional remote sensing image analysis and recognition mode, the image recognition mode has lower cost, and simultaneously effectively utilizes the position information of the internet data and the image text information, and improves the recognition accuracy through multi-source fusion mutual authentication; the accuracy of the photovoltaic map is further improved by adopting the actual photovoltaic facility data of the photovoltaic energy management platform, and the processing efficiency is improved.
The deep learning model is recommended to be adopted for paving the photovoltaic facilities, the internal relation existing between paved photovoltaic areas can be better found, the photovoltaic paved areas are identified, the area and the photovoltaic installed capacity are estimated by combining illumination information, and effective support is provided for accurate planning. In addition, based on multisource data, photovoltaic facility data updated in real time and external platform docking data, personalized data display is achieved based on a photovoltaic map, a model is continuously optimized, finer photovoltaic system operation is formed, and platform operation efficiency is improved.
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 required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are 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 schematic flow chart of a method for generating a multi-source information photovoltaic map.
Detailed Description
In order to provide a better understanding of the aspects of the present invention, the present invention will be described in further detail with reference to specific embodiments. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A preferred embodiment is given below:
As shown in fig. 1, in the method for generating a multi-source information photovoltaic map in this embodiment, correlation between multiple bands of remote sensing data is fully considered, a photovoltaic facility identification model and a photovoltaic facility laying recommendation model are designed based on a remote sensing image top view in combination with actual operation data of a photovoltaic energy management platform, and photovoltaic laying query analysis results are obtained by querying data related to photovoltaic laying including images from the internet and utilizing an image text photovoltaic analysis identification model.
And (3) performing target identification and mutual cross verification by integrating multiparty data, and constructing a photovoltaic map based on the remote sensing image map, the additional real-time operation data and the photovoltaic paving recommendation to form more accurate photovoltaic system operation and planning.
The remote sensing image is an image picture generated by image fusion of a multispectral image and a full-color image which are obtained through satellite shooting; the photovoltaic energy management platform is deployed in a cloud data center, provides functions of real-time data acquisition, asset management, instruction control, OTA upgrading, information display and the like of photovoltaic facilities, and provides infrastructure services such as calculation, storage, network and the like for the cloud data center to meet the resource requirements required by the operation of the photovoltaic energy management platform, and simultaneously provides Internet of things service, database service, AI service, deep learning model training and the like.
The core of the photovoltaic facility identification model RS-PVI-Det is a target detection model of the CNN convolutional neural network, and a target area of the photovoltaic facility is identified by inputting a remote sensing image for target detection. The core of the photovoltaic facility paving recommendation model RS-PVI-Pred is a target detection model based on a CNN convolutional neural network, and based on remote sensing images, the region suitable for photovoltaic paving and the degree suitable for paving the photovoltaic facilities are output.
The internet photovoltaic laying related information is the photovoltaic laying related data containing images of digital media such as microblogs, blogs, news and the like from the internet, and the information is obtained through searching by an internet photovoltaic information searching module. The internet photovoltaic information inquiry module IE-PVI-Search obtains the related information of internet photovoltaic laying by means of a digital media Search engine based on the position coordinate information or digital media inquiry of photovoltaic laying text keywords such as microblogs, blogs, news and the like.
The IMG-PVI-Det core of the image text photovoltaic analysis recognition model is a network model based on text recognition and image recognition and performing comprehensive analysis, semantic data in the network model are extracted, and comprehensive analysis recognition is performed by combining an image recognition result to obtain a photovoltaic laying inquiry analysis result.
And the Cross-Check model is based on the output result from the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det based on the same coordinate position area, and a mode of combining a preset rule with a neural network is adopted to judge the formed judging result.
The photovoltaic Map generation module PVI-Map-Gen is arranged on a remote sensing image Map, and is used for constructing the photovoltaic Map by adding operation data such as real-time acquisition data and power generation capacity of the photovoltaic facilities based on the identification result and the data of the actual photovoltaic facilities managed by the photovoltaic energy management platform, and providing recommended suggestion of a photovoltaic laying area and displaying planning prediction data such as expected yield and power generation capacity.
Training for photovoltaic facility identification models includes:
S101, collecting remote sensing image data and marking photovoltaic facilities;
S102, designing a photovoltaic facility identification model RS-PVI-Det based on a target detection model;
S103, performing image cutting on remote sensing image data according to the input size of the photovoltaic facility identification model RS-PVI-Det;
s104, training the photovoltaic facility identification model RS-PVI-Det to form the photovoltaic facility identification model RS-PVI-Det.
Training of the photovoltaic facility paving recommendation model includes:
s201, collecting related data of photovoltaic facilities of actual planning operation provided by a photovoltaic energy management platform;
S202, marking remote sensing images before paving the actually operated photovoltaic facilities provided by the photovoltaic energy management platform based on actual paving conditions;
S203, marking data of a photovoltaic facility area which is provided by the photovoltaic energy management platform and is planned and built on a remote sensing image;
s204, designing a photovoltaic facility paving recommendation model RS-PVI-Pred;
s205, performing image cutting on remote sensing image data according to the input size of a RS-PVI-Pred of a photovoltaic facility paving recommendation model;
S206, training the photovoltaic facility paving recommendation model RS-PVI-Pred to form the photovoltaic facility paving recommendation model RS-PVI-Pred.
Training of the image text photovoltaic analysis recognition model comprises the following steps:
S301, collecting relevant data about photovoltaic, roof, top view, unmanned aerial vehicle and the like with coordinate position data on digital media such as microblogs, blogs, news and the like from the Internet through an internet photovoltaic information inquiry module IE-PVI-Search;
s302, cleaning and filtering grouping is carried out on the collected data, and data marking is carried out;
S303, designing an image character photovoltaic analysis and identification model IMG-PVI-Det;
S304, the combination training is carried out on the model by taking the marked images and characters and combining the data such as coordinates, shooting date, search keywords and the like as input, so as to form an image character photovoltaic analysis recognition model IMG-PVI-Det.
The cross-validation model training includes:
S401, forming a basic input data set by the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det based on the output result of the same coordinate position area based on remote sensing image data and internet digital media data;
S402, performing cross-validation result judgment data annotation on a basic input data set to form a data set;
s403, designing the Cross-Check model;
S404, fixing network parameters of the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det, and performing joint training on a Cross-Check model;
S405, extracting set business rules according to actual photovoltaic facility data of the photovoltaic energy platform, and combining the Cross-Check model of the combined training to form the Cross-Check model of the combined training.
The steps in generating the photovoltaic map include:
S501, setting a map generation area position range;
s502, identifying a photovoltaic facility target area through the photovoltaic facility identification model RS-PVI-Det based on remote sensing image data and a set area position range;
S503, searching digital media such as microblogs, blogs, news and the like in a set area through an internet photovoltaic information inquiry module IE-PVI-Search in the remote sensing image data and the set area range, and analyzing and identifying based on the image text photovoltaic analysis and identification model IMG-PVI-Det to obtain results such as a target image area, geographic position coordinates, semantics, confidence and the like;
S504, judging the result output by the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det by combining the actual photovoltaic facility data of the photovoltaic energy platform and the data in the same area through the Cross-verification model Cross-Check;
s505, outputting a recommended photovoltaic facility paving region and a recommended degree value by using the photovoltaic facility paving recommended model RS-PVI-Pred based on remote sensing image data and a set region position range;
s506, according to the output from the step S502 to the step S505, generating a photovoltaic map based on the remote sensing image data and the position data and combining the actual photovoltaic facility data and the planning data of the photovoltaic energy platform.
Application and optimization of a photovoltaic map, comprising:
s601, acquiring data of photovoltaic facilities in real time by using a photovoltaic energy platform, and attaching the data to a map for display and continuously updating;
S602, collecting data of photovoltaic facilities in real time by a photovoltaic energy platform, adding the data to a map for display, distinguishing the data by colors, and providing related estimation data for decision planning;
s603, docking an external photovoltaic energy platform, accessing data to the platform, attaching the data to a photovoltaic map according to the position data for display, and distinguishing the data by colors;
S604, according to different attention points, personalized data display is carried out, information such as roof sunshine and shadow states, photovoltaic resources and development mode selection is displayed, actual feedback is collected, a model is continuously optimized, and a more reasonable and accurate photovoltaic map is generated.
Based on the method, the multi-source information photovoltaic map generation device comprises the following steps: at least one memory and at least one processor;
the at least one memory for storing a machine readable program;
The at least one processor is configured to invoke the machine-readable program to perform a multi-source information photovoltaic map generation method.
The above specific embodiments are merely examples of the present invention, and the scope of the present invention includes, but is not limited to, the above specific embodiments, any suitable changes or substitutions made by one of ordinary skill in the art, which are consistent with the present invention, of a multi-source information photovoltaic map generating method and apparatus claims, and all fall within the scope of the present invention.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. The method is characterized in that correlation among multiple bands of remote sensing data is utilized, actual operation data of a photovoltaic energy management platform is combined, a photovoltaic facility identification model and a photovoltaic facility paving recommendation model are designed based on a remote sensing image top view, photovoltaic paving related data from an Internet image is inquired, and a photovoltaic paving inquiry analysis result is obtained by utilizing an image text photovoltaic analysis identification model;
Target identification and mutual cross verification are carried out by integrating multiparty data, a photovoltaic map is constructed based on the remote sensing image map, the real-time operation data and the photovoltaic paving recommendation, and more accurate photovoltaic system operation and planning are formed;
The remote sensing image is an image picture generated by image fusion of a multispectral image and a full-color image which are obtained through satellite shooting; the photovoltaic energy management platform is deployed in a cloud data center for providing photovoltaic facility data acquisition, and the cloud data center meets the resource requirements required by the operation of the photovoltaic energy management platform;
the core of the photovoltaic facility identification model RS-PVI-Det is a target detection model of a CNN convolutional neural network, and a photovoltaic facility target area is identified by inputting a remote sensing image for target detection;
The core of the photovoltaic facility paving recommendation model RS-PVI-Pred is a target detection model based on a CNN convolutional neural network, and based on remote sensing images, the region suitable for photovoltaic paving and the degree suitable for paving the photovoltaic facilities are output;
The IMG-PVI-Det core is a network model based on text recognition and image recognition and performing comprehensive analysis, semantic data in the network model are extracted, and comprehensive analysis and recognition are performed by combining the image recognition result to obtain a photovoltaic laying inquiry analysis result;
cross-verifying to obtain a Cross-verifying model, wherein the Cross-Check is based on output results from the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det based on the same coordinate position area, and a judgment result formed by judging the Cross-verifying model by adopting a mode of combining a preset rule with a neural network;
The photovoltaic Map generation module PVI-Map-Gen is arranged on a remote sensing image Map, and is used for constructing the photovoltaic Map by adding real-time acquisition data and generating capacity operation data of the photovoltaic facilities based on the identification result and the data of the actual photovoltaic facilities managed by the photovoltaic energy management platform, providing recommendation suggestions of a photovoltaic paving area and displaying expected income and generating capacity planning prediction data;
The steps in generating the photovoltaic map include:
S501, setting a map generation area position range;
s502, identifying a photovoltaic facility target area through the photovoltaic facility identification model RS-PVI-Det based on remote sensing image data and a set area position range;
S503, searching digital media in a set area through an internet photovoltaic information inquiry module IE-PVI-Search in the remote sensing image data and the set area range, and analyzing and identifying based on the image text photovoltaic analysis and identification model IMG-PVI-Det to obtain a target result;
S504, judging the result output by the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det by combining the actual photovoltaic facility data of the photovoltaic energy platform and the data in the same area through the Cross-verification model Cross-Check;
s505, outputting a recommended photovoltaic facility paving region and a recommended degree value by using the photovoltaic facility paving recommended model RS-PVI-Pred based on remote sensing image data and a set region position range;
s506, according to the output from the step S502 to the step S505, generating a photovoltaic map based on the remote sensing image data and the position data and combining the actual photovoltaic facility data and the planning data of the photovoltaic energy platform.
2. The method for generating a multi-source information photovoltaic map according to claim 1, wherein the training for the photovoltaic facility identification model comprises:
S101, collecting remote sensing image data and marking photovoltaic facilities;
S102, designing a photovoltaic facility identification model RS-PVI-Det based on a target detection model;
S103, performing image cutting on remote sensing image data according to the input size of the photovoltaic facility identification model RS-PVI-Det;
s104, training the photovoltaic facility identification model RS-PVI-Det to form the photovoltaic facility identification model RS-PVI-Det.
3. The method for generating a multi-source information photovoltaic map according to claim 2, wherein training of the photovoltaic facility laying recommendation model comprises:
s201, collecting related data of photovoltaic facilities of actual planning operation provided by a photovoltaic energy management platform;
S202, marking remote sensing images before paving the actually operated photovoltaic facilities provided by the photovoltaic energy management platform based on actual paving conditions;
S203, marking data of a photovoltaic facility area which is provided by the photovoltaic energy management platform and is planned and built on a remote sensing image;
s204, designing a photovoltaic facility paving recommendation model RS-PVI-Pred;
s205, performing image cutting on remote sensing image data according to the input size of a RS-PVI-Pred of a photovoltaic facility paving recommendation model;
S206, training the photovoltaic facility paving recommendation model RS-PVI-Pred to form the photovoltaic facility paving recommendation model RS-PVI-Pred.
4. A method of generating a multi-source information photovoltaic map according to claim 3, wherein training of the image text photovoltaic analysis recognition model comprises:
S301, collecting related data with coordinate position data on a digital medium through an internet photovoltaic information inquiry module IE-PVI-Search;
s302, cleaning and filtering grouping is carried out on the collected data, and data marking is carried out;
S303, designing an image character photovoltaic analysis and identification model IMG-PVI-Det;
S304, carrying out joint training on the model by taking the marked keyword data as input to form an image character photovoltaic analysis recognition model IMG-PVI-Det.
5. The method of generating a multi-source information photovoltaic map according to claim 4, wherein the cross-validation model training comprises:
S401, forming a basic input data set by the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det based on the output result of the same coordinate position area based on remote sensing image data and internet digital media data;
S402, performing cross-validation result judgment data annotation on a basic input data set to form a data set;
s403, designing the Cross-Check model;
S404, fixing network parameters of the photovoltaic facility identification model RS-PVI-Det and the image text photovoltaic analysis identification model IMG-PVI-Det, and performing joint training on a Cross-Check model;
S405, extracting set business rules according to actual photovoltaic facility data of the photovoltaic energy platform, and combining the Cross-Check model of the combined training to form the Cross-Check model of the combined training.
6. The method for generating a multi-source information photovoltaic map according to claim 5, wherein the application and optimization of the photovoltaic map comprises:
s601, acquiring data of photovoltaic facilities in real time by using a photovoltaic energy platform, and attaching the data to a map for display and continuously updating;
S602, collecting data of photovoltaic facilities in real time by a photovoltaic energy platform, adding the data to a map for display, distinguishing the data by colors, and providing related estimation data for decision planning;
s603, docking an external photovoltaic energy platform, accessing data to the platform, attaching the data to a photovoltaic map according to the position data for display, and distinguishing the data by colors;
S604, according to different attention points, personalized data display is carried out, roof sunshine and shadow states, photovoltaic resources and development mode selection information are displayed, actual feedback is collected, a model is continuously optimized, and a more reasonable and accurate photovoltaic map is generated.
7. A multi-source information photovoltaic map generation apparatus, comprising: at least one memory and at least one processor;
the at least one memory for storing a machine readable program;
The at least one processor being configured to invoke the machine readable program to perform the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210063493.8A CN114494830B (en) | 2022-01-20 | 2022-01-20 | Multi-source information photovoltaic map generation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210063493.8A CN114494830B (en) | 2022-01-20 | 2022-01-20 | Multi-source information photovoltaic map generation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114494830A CN114494830A (en) | 2022-05-13 |
| CN114494830B true CN114494830B (en) | 2024-05-17 |
Family
ID=81471718
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210063493.8A Active CN114494830B (en) | 2022-01-20 | 2022-01-20 | Multi-source information photovoltaic map generation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114494830B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114896437B (en) * | 2022-07-14 | 2022-09-13 | 北京数慧时空信息技术有限公司 | Remote sensing image recommendation method based on available domain |
| CN116188585B (en) * | 2023-04-24 | 2023-07-11 | 成都垣景科技有限公司 | Mountain area photovoltaic target positioning method based on unmanned aerial vehicle photogrammetry |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111191500A (en) * | 2019-11-12 | 2020-05-22 | 广东融合通信股份有限公司 | Photovoltaic roof resource identification method based on deep learning image segmentation |
| CN111611334A (en) * | 2020-04-24 | 2020-09-01 | 国家电网有限公司 | Power grid geographic information system fusing multi-source information |
| CN113128793A (en) * | 2021-05-19 | 2021-07-16 | 中国南方电网有限责任公司 | Photovoltaic power combination prediction method and system based on multi-source data fusion |
| WO2021244000A1 (en) * | 2020-06-03 | 2021-12-09 | 国网上海市电力公司 | Virtual aggregation system and method for regional energy source complex |
-
2022
- 2022-01-20 CN CN202210063493.8A patent/CN114494830B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111191500A (en) * | 2019-11-12 | 2020-05-22 | 广东融合通信股份有限公司 | Photovoltaic roof resource identification method based on deep learning image segmentation |
| CN111611334A (en) * | 2020-04-24 | 2020-09-01 | 国家电网有限公司 | Power grid geographic information system fusing multi-source information |
| WO2021244000A1 (en) * | 2020-06-03 | 2021-12-09 | 国网上海市电力公司 | Virtual aggregation system and method for regional energy source complex |
| CN113128793A (en) * | 2021-05-19 | 2021-07-16 | 中国南方电网有限责任公司 | Photovoltaic power combination prediction method and system based on multi-source data fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114494830A (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114494830B (en) | Multi-source information photovoltaic map generation method and device | |
| CN107273530B (en) | Internet information-based important ship target dynamic monitoring method | |
| US10768769B2 (en) | Automatic event detection, text generation, and use thereof | |
| Niu et al. | Real-time taxi-passenger prediction with L-CNN | |
| Xu et al. | A dynamic topic model and matrix factorization-based travel recommendation method exploiting ubiquitous data | |
| CN101842788A (en) | Method, apparatus and computer program product for performing a visual search using grid-based feature organization | |
| Song et al. | Detection of oil wells based on faster R-CNN in optical satellite remote sensing images | |
| Nasrallah et al. | Lebanon solar rooftop potential assessment using buildings segmentation from aerial images | |
| He et al. | Perceiving commerial activeness over satellite images | |
| Fan et al. | DuMapper: Towards Automatic Verification of Large-Scale POIs with Street Views at Baidu Maps | |
| CN114529721B (en) | Urban remote sensing image vegetation coverage recognition method based on deep learning | |
| JP2013045182A (en) | Information retrieval apparatus, method, and program | |
| Miao et al. | TransLinkNet: LinkNet with transformer for road extraction | |
| Hao et al. | UrbanVLP: A Multi-Granularity Vision-Language Pre-Trained Foundation Model for Urban Indicator Prediction | |
| JP2022088574A (en) | Map query method, device, and electronic apparatus | |
| Bayram et al. | A Deep learning integrated mobile application for historic landmark recognition: A case study of Istanbul | |
| Ishikawa et al. | Towards construction of an explanation framework for whole processes of data analysis applications: concepts and use cases | |
| Afif et al. | Indoor objects detection system implementation using multi-graphic processing units | |
| Sun et al. | A cross-view geo-localization method guided by relation-aware global attention | |
| Tang et al. | Research on the Network Map Service Technology of Remote Sensing Image Intelligent Conversion Based on Gan Model | |
| Guo et al. | Remote sensing identification of green plastic cover in urban built-up areas | |
| Du et al. | Mining multicity urban data for sustainable population relocation | |
| CN110837780A (en) | Image interpretation hierarchical and capability modular training method | |
| Qian et al. | Multi-task deep learning for large-scale building detail extraction from high-resolution satellite imagery | |
| Brijith | Smart City: An AI-based disruptive revolution |
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 |