CN116957223B - Photovoltaic power prediction method and device based on illumination image - Google Patents
Photovoltaic power prediction method and device based on illumination image Download PDFInfo
- Publication number
- CN116957223B CN116957223B CN202310708777.2A CN202310708777A CN116957223B CN 116957223 B CN116957223 B CN 116957223B CN 202310708777 A CN202310708777 A CN 202310708777A CN 116957223 B CN116957223 B CN 116957223B
- Authority
- CN
- China
- Prior art keywords
- illumination
- image
- illumination intensity
- images
- outdoor
- 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
- 238000005286 illumination Methods 0.000 title claims abstract description 240
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 238000001514 detection method Methods 0.000 claims description 27
- 230000000737 periodic effect Effects 0.000 claims description 23
- 238000010248 power generation Methods 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 230000002060 circadian Effects 0.000 claims description 7
- 230000001932 seasonal effect Effects 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 abstract description 8
- 230000008859 change Effects 0.000 description 20
- 230000006870 function Effects 0.000 description 18
- 230000008569 process Effects 0.000 description 8
- 230000002829 reductive effect Effects 0.000 description 6
- 238000004590 computer program Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000001965 increasing effect Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 235000019800 disodium phosphate Nutrition 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000000670 limiting effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 241000112598 Pseudoblennius percoides Species 0.000 description 1
- 238000013531 bayesian neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003867 tiredness Effects 0.000 description 1
- 208000016255 tiredness Diseases 0.000 description 1
- 230000001052 transient effect 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
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/35—Categorising the entire scene, e.g. birthday party or wedding scene
- G06V20/38—Outdoor scenes
-
- 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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Power Engineering (AREA)
- Operations Research (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Photovoltaic Devices (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of photovoltaics, in particular to a photovoltaic power prediction method and device based on illumination images. In the technical scheme provided by the embodiment of the application, the photovoltaic power in a specific period is directly obtained based on the prediction of the most critical illumination intensity characteristic in the photovoltaic power influence. The prediction of the illumination intensity characteristic is based on the illumination image acquired in the period, the key association factors in the illumination image are predicted based on the neural network, the final illumination intensity characteristic is obtained based on the key association factors, the photovoltaic power predicted value is directly obtained based on the predicted illumination intensity, and the problem of huge training data caused by the arrangement of the large-scale neural network in the prior art is solved while the prediction accuracy is ensured.
Description
Technical Field
The invention relates to the technical field of photovoltaics, in particular to a photovoltaic power prediction method and device based on an illumination image.
Background
In recent years, the photovoltaic power generation of China rapidly develops, and the installed capacity of the photovoltaic power is increased year by year. By the end of 2021, the installed capacity of the photovoltaic power generation grid-connected system in China reaches 3.06 hundred million kilowatts, the newly installed capacity in the current year is about 5300 kilowatts, and the same ratio is increased by about 21%. The great development of new energy sources such as photovoltaics has become an important measure for promoting energy transformation. However, the photovoltaic output is greatly affected by weather, and the photovoltaic power generation system has the characteristics of obvious nonlinearity, volatility, uncertainty and the like, and the large-scale access to the power grid inevitably brings challenges to the safe and stable operation of the power grid. Therefore, the method and the device have important significance in timely and accurate prediction of photovoltaic power, optimal scheduling of a power grid, economic operation of a photovoltaic power station and the like.
In the prior art, the photovoltaic prediction scheme is mainly based on a complex neural network to predict by acquiring external environment changes, and has higher accuracy in the areas with larger external climate environment changes. However, because the data set and the data characteristics required for training the complex model are large, the training process for the model is complex, and the deployment of the model has practical significance for use because of high accuracy for the region with large external climate environment change. But the deployment is less significant and has greater cost for areas where the external climate environment is not changing much or is not intense. Therefore, there is a need to provide a photovoltaic power prediction method based on illumination images with high accuracy and low deployment cost for scenes where the actual climate region has variations but the variations are not severe.
Disclosure of Invention
In order to achieve the technical effects, the photovoltaic power prediction method based on the photovoltaic power main influence factors is used as the prediction features, a neural network model with accurate feature prediction is built, and the photovoltaic power prediction method under the condition of non-severe climate change is achieved through the method, so that the cost of model deployment and the cost of prediction are reduced on the basis of high accuracy.
In order to achieve the above purpose, the technical solution adopted in the embodiment of the present application is as follows:
in a first aspect, a photovoltaic power prediction method based on illumination images is applied to a server, and the method includes: collecting a plurality of outdoor illumination images corresponding to a photovoltaic system according to a periodic collection strategy, and screening the collected outdoor illumination images to obtain a target illumination image, wherein the periodic collection strategy is set based on periodic changes of illumination intensity and comprises circadian periodic changes and seasonal periodic changes, the periodic collection strategy comprises a collection sub-strategy, and the collection sub-strategy comprises at least three times of outdoor illumination image collection; inputting the target illumination image into a trained illumination intensity detection model to obtain illumination intensity estimation; and obtaining a power coefficient based on the illumination intensity estimation and the standard illumination intensity, and obtaining a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
Further, collecting a plurality of outdoor illumination images corresponding to the photovoltaic system according to a periodic collection strategy, including: and acquiring outdoor illumination images in the corresponding season sections according to the seasonal variation, and acquiring a plurality of outdoor illumination images in the corresponding unit time sections according to the diurnal variation.
Further, acquiring a plurality of outdoor illumination images within a corresponding unit time period according to the circadian variation includes: and acquiring at least three outdoor illumination images in a corresponding unit time period according to the circadian cycle variation.
Further, screening the collected outdoor illumination images to obtain a target illumination image, including: acquiring the gray scale of at least three outdoor illumination images, acquiring at least three gray scale ratio values among the at least three outdoor illumination images, and taking the finally acquired outdoor illumination images as target illumination images when the gray scale ratio values are within a preset threshold range; and when the gray ratio is not in the preset threshold range, eliminating the outdoor illumination image corresponding to the ratio exceeding the preset threshold range, and taking the last acquired image in the rest outdoor illumination images as a target illumination image, wherein the target illumination image is an RGB image.
Further, before the target illumination image is input to the trained illumination intensity detection model, color space conversion processing is further included on the target illumination image, the target illumination image is converted into HSV color space, and three channels are respectively output, wherein the value of the H channel is the hue of the color, the value of the S channel is the saturation of the color, and the value of the V channel is the brightness value of the color.
Further, extracting the value of the V channel is included before inputting the target illumination image into the trained illumination intensity detection model.
Further, inputting the target illumination image to a trained illumination intensity detection model includes inputting the target illumination image and the V-channel value to the trained illumination intensity detection model.
Further, the illumination intensity detection model is a neural network structure model and comprises an input layer, a convolution layer and a 9-layer residual pouring module connected with the convolution layer, wherein the residual pouring module is connected with two full-connection layers, and the two full-connection layers are respectively provided with an activation function and batch normalization operation and respectively output the probability distribution of the sun direction and the atmospheric turbidity; the method for obtaining the illumination intensity estimation comprises the steps of: inputting the target illumination image into a trained illumination intensity detection model, and outputting probability distribution and atmospheric turbidity of the sun direction corresponding to the target illumination image by the illumination intensity detection model; and obtaining illumination intensity estimation based on the probability distribution of the sun direction, the atmospheric turbidity and a sun illumination constant, and determining the illumination intensity estimation.
Further, the solar light constant is 2.838X10 27 cd。
In a second aspect, there is provided a photovoltaic power prediction apparatus based on an illumination image, the apparatus comprising: the outdoor illumination image acquisition module is used for acquiring a plurality of outdoor illumination images corresponding to the photovoltaic system according to a periodic acquisition strategy and screening the acquired outdoor illumination images to obtain a target illumination image; the illumination intensity estimation module is used for acquiring illumination intensity estimation based on the target illumination image; and the photovoltaic power prediction module is used for obtaining a power coefficient based on the illumination intensity estimation and the standard illumination intensity and obtaining a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
In a third aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program which, when executed by a processor, implements the method of any one of the above.
In the technical scheme provided by the embodiment of the application, the photovoltaic power in a specific period is directly obtained based on the prediction of the most critical illumination intensity characteristic in the photovoltaic power influence. The prediction of the illumination intensity characteristic is based on the illumination image acquired in the period, the key association factors in the illumination image are predicted based on the neural network, the final illumination intensity characteristic is obtained based on the key association factors, the photovoltaic power predicted value is directly obtained based on the predicted illumination intensity, and the problem of huge training data caused by the arrangement of the large-scale neural network in the prior art is solved while the prediction accuracy is ensured.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
The methods, systems, and/or programs in the accompanying drawings will be described further in terms of exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein the exemplary numbers represent like mechanisms throughout the various views of the drawings.
Fig. 1 is a schematic flow chart of a photovoltaic power prediction method based on illumination images according to an embodiment of the present application.
Fig. 2 is a block diagram of a photovoltaic power prediction apparatus based on illumination images according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of a photovoltaic power prediction device based on illumination images according to an embodiment of the present application.
Detailed Description
In order to better understand the technical solutions described above, the following detailed description of the technical solutions of the present application is provided through the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and embodiments of the present application are detailed descriptions of the technical solutions of the present application, and not limit the technical solutions of the present application, and the technical features of the embodiments and embodiments of the present application may be combined with each other without conflict.
In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant teachings. However, it will be apparent to one skilled in the art that the present application may be practiced without these details. In other instances, well-known methods, procedures, systems, components, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.
The flowcharts are used in this application to describe implementations performed by systems according to embodiments of the present application. It should be clearly understood that the execution of the flowcharts may be performed out of order. Rather, these implementations may be performed in reverse order or concurrently. Additionally, at least one other execution may be added to the flowchart. One or more of the executions may be deleted from the flowchart.
Before describing embodiments of the present invention in further detail, the terms and terminology involved in the embodiments of the present invention will be described, and the terms and terminology involved in the embodiments of the present invention will be used in the following explanation.
(1) In response to a condition or state that is used to represent the condition or state upon which the performed operation depends, the performed operation or operations may be in real-time or with a set delay when the condition or state upon which it depends is satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.
(2) Based on the conditions or states that are used to represent the operations that are being performed, one or more of the operations that are being performed may be in real-time or with a set delay when the conditions or states that are being relied upon are satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.
(3) Neural networks are a class of feedforward neural networks that involve convolution calculations and have a deep structure. Convolutional neural networks are proposed by biological Receptive Field (fielded) mechanisms. Convolutional neural networks are dedicated to neural networks that process data having a grid-like structure. For example, time-series data (which may be regarded as a one-dimensional grid formed by regularly sampling on a time axis) and image data (which may be regarded as a two-dimensional grid of pixels), the convolutional neural network employed in the present embodiment processes, with respect to the image data, that is, a graph neural network (GCN).
The shortage of energy is considered to be an increasingly serious problem, restricts the sustainable development of economy and society, and all countries should develop renewable energy without surplus energy, reduce the consumption of traditional energy, promote the energy transformation development of the international society and cope with uncertain climate change. Under the current situation of existing energy, the dependence on fossil energy needs to be reduced, the proportion of thermal power to coal power is reduced, and new energy is developed to replace the traditional fossil energy. The new energy source comprises solar energy, wind energy, tidal energy, geothermal energy and the like, wherein the solar energy is used as a renewable clean energy source which is easy to obtain and has huge resource potential. Photovoltaic power generation is a main mode of solar energy application, and in recent years, a large-scale photovoltaic power station is connected into an intelligent power distribution network, so that the photovoltaic power generation technology becomes an important component of the intelligent power distribution network continuously, and is beneficial to meeting the energy demand of peak time, and has a good development prospect.
With the global integration of photovoltaic power and the increasing installed capacity, power system operators face many challenges, one of the key challenges for power integration being its randomness, volatility and uncertainty. Since the output of solar photovoltaic power generation is mainly dependent on weather, weather conditions always control the output of the photovoltaic system, making the output power variable, its main influencing factors are: irradiation intensity, cloud cover, temperature, humidity, and the like. Thus, unlike conventional generator outputs, which can be adjusted to a desired value, the photovoltaic power output is not controllable, and its stability and safety will be severely compromised when the grid is connected to a large solar power plant. In addition, the high permeability problem caused by large-scale photovoltaic grid connection increases the complexity and the control difficulty of the power grid, and the large-scale distributed photovoltaic multipoint irregular access to the power distribution network causes the problems of power quality, harmonic waves, economic operation and the like for the power grid. The effective way of fully utilizing the photovoltaic power generation system is to keep the total output power of the multi-energy system relatively stable, reduce power fluctuation, improve the power quality and reduce the influence on a power grid.
The current situation and development trend of photovoltaic power generation at present are that authorities are prompted to effectively manage a power grid, enough and stable power resources are provided for users, short-term prediction of solar photovoltaic power generation power is crucial for reliable and safe operation of the power grid, and the significance of the short-term prediction of photovoltaic power mainly comprises the following points: 1) Aiming at the fact that the fluctuation and randomness of the photovoltaic output power are large, the transportation consumption and the unstable voltage are caused, the accurate short-term photovoltaic power generation prediction is uniquely helpful for balancing the supply and the demand of a power system, the economic cost can be reduced, and the method has important significance for the economic dispatching and the optimized operation of the power system and the photovoltaic power generation system. 2) The power generated by the photovoltaic system in the photovoltaic grid connection is affected by the environment and is continuously changed, from the aspects of scheduled needs and safety management of the power grid, the accurate photovoltaic power prediction can enable the power grid to reduce the problem of light abandoning and electricity limiting, and the permeability of the photovoltaic grid connection is improved by ensuring the dynamic balance of the system. 3) The working personnel can select a low-output time period in advance one day according to the result of photovoltaic power prediction, and the photovoltaic unit equipment is subjected to shutdown maintenance and daily maintenance in the corresponding time period, so that the possibility of fault occurrence is reduced, and the running reliability of the whole power system is improved.
Photovoltaic probabilistic predictive models are classified into physical, statistical, and artificial intelligence probabilistic predictive models. The physical probability prediction model is characterized in that certain parameter distribution is selected according to experience to simulate irradiance probability distribution (such as Normal, weibull and Beta distribution, and the like), then a probability prediction model of photovoltaic output power is established according to a functional relation between photovoltaic power and irradiance, the probability prediction model is suitable for a photovoltaic power prediction scene considering factors such as geographic conditions, ground environment, and the like, a prediction model for combined statistical description of solar photovoltaic power generation power and outdoor temperature is provided in the prior art, potential correlation between temperature and solar photovoltaic power can be well described only by a simple low-complexity sampling space, probability prediction results are obtained through the combined model, and the probability prediction model has better application value compared with deterministic point prediction;
the statistical probability prediction model does not need to master internal information of a photovoltaic power station system, is a data driving method for predicting future output power by using NWP data and photovoltaic power historical data, generally depends on a large number of historical samples, and the common statistical probability prediction model comprises nuclear density estimation, quantile regression and the like, a pseudo-data self-adaptive nuclear density estimation probability prediction model is built in the prior art, a pseudo-data method is adopted for correcting boundary deviation of nuclear density estimation, local adaptability can be improved, and the power load interval prediction model combining quantile regression average and sister point prediction is provided in the prior art, so that the method has higher practical value.
The artificial intelligent probabilistic predictive model can fully mine the complex nonlinear relation between input data and output data and has the characteristics of simple algorithm realization, strong generalization capability and the like. The common artificial intelligence probability prediction model is Bayesian learning, neural network and the like. Firstly, performing point prediction by adopting a machine learning or deep learning model, obtaining a probability prediction result by analyzing statistical characteristics of prediction errors, combining a neural network and non-parameter probability prediction to perform interval prediction in the prior art, quantifying uncertainty of a photovoltaic power prediction result, providing a prediction method based on convolution neural network and gate control circulation neural network quantile regression in the prior art, extracting characteristic information by the convolution neural network, taking the characteristic information as input of a gate control circulation neural network quantile regression model, obtaining prediction results of different quantiles, and obtaining probability density distribution by adopting Gaussian kernel density estimation, thereby solving the problem of power load probability density prediction; the other is to directly predict and obtain the probability prediction result by using the loss function, such as a photovoltaic output probability prediction method based on an improved Bayesian neural network in the prior art, and directly obtain the upper and lower limit prediction values of the probability interval.
And for the prediction model of the physical class and the statistical class, the prediction result depends on large-scale statistical data, and the accuracy of the prediction result is lower than that of the scheme of the artificial intelligent model. The scheme aiming at the artificial intelligent model is based on a neural network to realize prediction, and because the directions of the main model solution at present are all under the condition of complex climate change, the model construction and training need to rely on a large amount of environmental data, and the requirement on the quantity of characteristic parameters is larger, so that the method is suitable for areas with more sensitive climate change, namely larger climate change can be generated in a short time. And the cost of arranging the artificial intelligent model is high for areas where climate change is not sensitive. Therefore, the arrangement of the model is not facilitated.
For the above background information, the embodiment of the application provides a photovoltaic power prediction method based on an illumination image, which specifically includes the following steps:
s110, acquiring a plurality of outdoor illumination images corresponding to the photovoltaic system according to a periodic acquisition strategy, and screening the acquired outdoor illumination images to obtain a target illumination image.
In the embodiment of the application, the periodic acquisition strategy is set based on the periodic variation of the illumination intensity. The periodic variation of illumination intensity is divided into seasonal variation and diurnal variation, wherein the time and intensity of the seasonal variation for illumination in different seasons are different, so that different acquisition schemes are required for the acquisition of outdoor illumination images for seasons. For example, when the illumination time in summer is longer, the acquisition strategy is different from that when the illumination time in winter is shorter, namely, the outdoor illumination image acquisition time in summer is longer than that in winter. And aiming at the circadian cycle change, as the sun rises and falls, acquisition schemes corresponding to different time periods need to be configured. Therefore, outdoor illumination images in the corresponding season segments are acquired according to the seasonal variation for the specific positions of the periodic acquisition strategy, and a plurality of outdoor illumination images in the corresponding unit time periods are acquired according to the diurnal variation.
Specifically, in the embodiment of the present application, the number of the plurality of outdoor illumination images for a unit time period is at least three, wherein the number is configured to be at least three because at least three images can characterize the change of different time periods when the images are processed later, and the optimal images can be obtained through the at least three images as the basis of the subsequent processing data.
By configuring the acquisition strategy, the optimal image acquisition scheme can be set based on natural change, and the problems of inaccurate image acquisition result and incomplete data caused by natural change are solved.
And aiming at the obtained at least three outdoor illumination images, a final target illumination image is required to be obtained in a screening mode, and the target illumination image is used as a basic image for subsequent processing. The screening method for the target illumination image is to acquire the gray scale of at least three outdoor illumination images, and determine the final target illumination image based on the relation of gray scale among the images. The gray scale of the image is used for representing the brightness condition of the environment where the image is located, the gray scale of the image collected at three different time points can be used for describing the brightness change of the environment where the image is located in different time periods, and the illumination change intensity of the environment where the photovoltaic system is located can be determined according to the brightness change of the image in different time periods.
The specific processing steps are as follows: and acquiring at least three corresponding gray scale ratios among the at least three outdoor illumination images, and when the gray scale ratios are within a preset threshold range, indicating that the brightness change belongs to the standard change meeting the expectation, and taking the finally acquired outdoor illumination image as a target illumination image according to the condition.
If the gray ratio is not within the preset threshold range, the outdoor illumination image corresponding to the ratio exceeding the preset threshold range is removed, and the last acquired image in the rest outdoor illumination images is used as a target illumination image, wherein the format of the target illumination image is RGB image. It is noted that, the photovoltaic power prediction method based on the illumination image provided by the embodiment of the application is mainly used for power prediction in a certain period, and is not used as an influence factor of prediction on occasional fluctuation, so that the purpose of obtaining the target illumination image is to determine an image which can be most characterized in the period change, and an outdoor illumination image corresponding to the occasional fluctuation is not used as a processed image. The accidental fluctuation is determined based on the gray ratio of the image in a plurality of time periods, and when the gray ratio is out of range, the accidental fluctuation is indicated to be eliminated.
And S120, inputting the target illumination image into a trained illumination intensity detection model to obtain illumination intensity estimation.
In the embodiment of the application, in order to improve the overall prediction accuracy, the input to the illumination intensity monitoring model not only includes the acquired target illumination image, but also includes the independent feature capable of representing the illumination intensity in the image. Therefore, before the model processing is performed on the target illumination image, the method further comprises the steps of obtaining independent features representing illumination intensity in the target illumination image, converting RGB of the target illumination image into HSV color space, and outputting three channels respectively. Wherein the value of the H channel is the hue of the color, the value of the S channel is the saturation of the color, and the value of the V channel is the brightness value of the color. Where the value of S channel is the saturation of the color, which may represent the shade of the color. The value of the V channel is the brightness value of the color, and may represent the brightness of the color. The values of the H channel and the S channel are hardly influenced by the ambient illumination, the value of the V channel is extremely sensitive to the change of the ambient illumination, the change of the illumination brightness can directly change the value of the V channel, and the characteristic can be used for enhancing the illumination information of the image. Therefore, for the embodiment of the application, the image is firstly converted into the HSV color space from the RGB color space, then the V channel is separated from the HSV color space and is used as a group of input of the illumination intensity detection model together with the original RGB three channels, so that the illumination characteristics of the image are enhanced, and the prediction precision of the model is improved. The specific method for converting the image color space and extracting the V-channel value may be an image processing method in the prior art, which is not described in detail in the embodiments of the present application.
The basic structure of the illumination intensity detection model in the embodiment of the application is a neural network structure, and the model comprises an input layer, a convolution layer and a 9-layer residual pouring module connected with the convolution layer, wherein the residual pouring module is connected with two full-connection layers, and the two full-connection layers are respectively configured with an activation function and batch normalization operation to respectively output the probability distribution of the sun direction and the atmospheric turbidity.
Specifically, for a convolution layer having a convolution kernel with a size of 3×3 and 16 output channels, an h-swish activation function is configured in the convolution layer. The structure of the residual pouring modules is a linear bottleneck structure, and attention mechanism modules are arranged between the residual pouring modules. Different filters and different numbers of output channels are configured in the inverted residual error modules of each layer, and the inverted residual error modules are divided into the following layers according to the sequence from the access convolution layer to the output full connection layer: 3 x 3 size filter, 5 x 5 size filter, and method for forming same wherein the activation functions corresponding to different levels are different, the method comprises the steps of a ReLU activation function, a h-swish activation function and a h-swish activation function respectively.
The probability distribution and the atmospheric turbidity about the solar direction are obtained through the model processing, wherein the maximum distribution in the probability distribution of the solar direction is taken as the final predicted solar direction, and the solar direction and the atmospheric turbidity are obtained to obtain the final illumination intensity estimation based on the solar illumination constant.
The dataset employed for training of the illumination intensity detection model in the embodiments of the present application is an image dataset with illumination parameter labels, wherein the construction of this dataset is based on the label dataset and the image dataset. Wherein the acquisition for the tag dataset and the image dataset is constructed by a simulation model, wherein the simulation model is a sky model. And acquiring sky panoramic image information aiming at the constructed chord of the image dataset and fitting the sky panoramic image information in a sky model to obtain four illumination parameters. Among the two parameters are azimuth and elevation angles of the sun, which can be used to represent a specific direction of sunlight. The third parameter is the atmospheric turbidity, which can directly affect the intensity of outdoor illumination. When the atmospheric turbidity is low, a brighter light environment may be indicated, whereas a darker light environment may be indicated. The last parameter is the elevation angle of the camera. And after the specific data of the four parameters are regulated to be arranged and rented, finally, a plurality of images can be obtained in total, and each image has four corresponding illumination parameters. Meanwhile, the tag dataset is also constructed in the same way. Finally, an image dataset with illumination parameter labels can be obtained.
In the training process of the network, two loss functions are used, one being a divergence loss function for estimating the solar direction distribution and the other being a mean square error loss function for the remaining two parameters. The network outputs two sets of parameters, the first set of parameters being a predicted probability distribution for the solar direction. The second set of parameters is that the predicted value q represents the atmospheric turbidity and the camera elevation angle, and p represents the ground-truthful atmospheric turbidity and the camera elevation angle. Since the solar direction is more important than the other two illumination parameters, the weights α=0.1 of the other two illumination parameters are set here. By using Adam optimizers, the value of the loss function can be minimized, with greater accuracy as the value of the loss function is smaller. In addition, the initial learning rate for the present model was 0.01, and the batch size for training was 64.
Wherein for solar illumination constants are solar illumination intensity data in the corresponding seasons and the corresponding time periods, the data can be statistical data, namely illumination intensity theoretical values under ideal conditions. In the embodiment of the application, the sun illumination constant is 2.838X10 27 cd。
Since the direct solar illumination intensity is known in a specific season and for a specific period of time, the illumination intensities corresponding to the sun for different angles can be obtained through simulation and through an angle conversion relation, and the direct solar illumination intensity can be realized by adopting the prior art, and no tiredness is caused. The atmospheric turbidity can be obtained by adopting a simulation mode, the part can also be realized by adopting the prior art, and the embodiment of the application is not tired.
The acquisition of the illumination intensity is based on a model, the logic of the acquisition is used for estimating the illumination intensity through image acquisition, and the physical data direct measurement is mainly carried out through setting a physical sensor aiming at the illumination intensity in the prior art. Compared with a physical sensor, the physical sensor is used for acquiring real-time sensing data, the prediction and estimation of illumination intensity cannot be realized, and the data acquisition is only real-time but cannot be predicted. And, the use cost for using the model may decrease as the use time increases, and the accuracy of the prediction result may be higher and higher because of the increase in the data amount. But the performance of the physical sensor is reduced due to the arrangement of physical components due to the increase of the use time, resulting in poor accuracy of the result.
And S130, obtaining a power coefficient based on the illumination intensity estimation and the standard illumination intensity, and obtaining a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
The power coefficient is obtained by dividing the illumination intensity estimation or the obtained illumination intensity estimation with the standard illumination intensity, and the final photovoltaic power prediction result is obtained by multiplying the power coefficient and the maximum power value realized by the photovoltaic theory.
Wherein the maximum power value is also a specific value known for the standard illumination intensity, and the acquisition of the above two data is not described in the embodiments of the present application.
Referring to fig. 2, there is provided a photovoltaic power prediction apparatus 200 based on illumination image, including:
the outdoor illumination image acquisition module 210 is configured to acquire a plurality of outdoor illumination images corresponding to the photovoltaic system according to a periodic acquisition strategy, and screen the acquired plurality of outdoor illumination images to obtain a target illumination image.
The illumination intensity estimation module 220 is configured to obtain an illumination intensity estimation based on the target illumination image.
The photovoltaic power prediction module 230 obtains a power coefficient based on the illumination intensity estimation and the standard illumination intensity, and obtains a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
Referring to fig. 3, the photovoltaic power prediction apparatus 300 may have a relatively large difference due to different configurations or performances, and may include one or more processors 301 and a memory 302, where the memory 302 may store one or more storage applications or data. Wherein the memory 302 may be transient storage or persistent storage. The application program stored in the memory 302 may include one or more modules (not shown in the figures), each of which may include a series of computer-executable instructions in the photovoltaic power prediction apparatus. Still further, the processor 401 may be arranged to communicate with the memory 302 and execute a series of computer executable instructions in the memory 302 on the photovoltaic power prediction device. The photovoltaic power prediction device may also include one or more power sources 303, one or more wired or wireless network interfaces 304, one or more input/output interfaces 305, one or more keyboards 306, and the like.
In one particular embodiment, a photovoltaic power prediction device includes a memory, and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions in the photovoltaic power prediction device, and configured to be executed by one or more processors, the one or more programs comprising computer-executable instructions for:
collecting a plurality of outdoor illumination images corresponding to a photovoltaic system according to a periodic collection strategy, and screening the collected outdoor illumination images to obtain a target illumination image;
inputting the target illumination image into a trained illumination intensity detection model to obtain illumination intensity estimation;
and obtaining a power coefficient based on the illumination intensity estimation and the standard illumination intensity, and obtaining a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
The following describes each component of the processor in detail:
wherein in the present embodiment, the processor is a specific integrated circuit (application specific integrated circuit, ASIC), or one or more integrated circuits configured to implement embodiments of the present application, such as: one or more microprocessors (digital signal processor, DSPs), or one or more field programmable gate arrays (field programmable gate array, FPGAs).
Alternatively, the processor may perform various functions, such as performing the method shown in fig. 2 described above, by running or executing a software program stored in memory, and invoking data stored in memory.
In a particular implementation, the processor may include one or more microprocessors, as one embodiment.
The memory is configured to store a software program for executing the solution of the present application, and the processor is used to control the execution of the software program, and the specific implementation manner may refer to the above method embodiment, which is not described herein again.
Alternatively, the memory may be read-only memory (ROM) or other type of static storage device that can store static information and instructions, random access memory (random access memory, RAM) or other type of dynamic storage device that can store information and instructions, but may also be, without limitation, electrically erasable programmable read-only memory (electrically erasable programmable read-only memory, EEPROM), compact disc read-only memory (compact disc read-only memory) or other optical disk storage, optical disk storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be integrated with the processor or may exist separately and be coupled to the processing unit through an interface circuit of the processor, which is not specifically limited in the embodiments of the present application.
It should be noted that the structure of the processor shown in this embodiment is not limited to the apparatus, and an actual apparatus may include more or less components than those shown in the drawings, or may combine some components, or may be different in arrangement of components.
In addition, the technical effects of the processor may refer to the technical effects of the method described in the foregoing method embodiments, which are not described herein.
It should be appreciated that the processor in embodiments of the present application may be other general purpose processors, digital signal processors (digital signal processor, DSP), application specific integrated circuits (application specific integrated circuit, ASIC), off-the-shelf programmable gate arrays (field programmable gate array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It should also be appreciated that the memory in embodiments of the present application may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The nonvolatile memory may be a read-only memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an electrically Erasable EPROM (EEPROM), or a flash memory. The volatile memory may be random access memory (random access memory, RAM) which acts as an external cache. By way of example but not limitation, many forms of random access memory (random access memory, RAM) are available, such as Static RAM (SRAM), dynamic Random Access Memory (DRAM), synchronous Dynamic Random Access Memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced Synchronous Dynamic Random Access Memory (ESDRAM), synchronous Link DRAM (SLDRAM), and direct memory bus RAM (DR RAM).
The above embodiments may be implemented in whole or in part by software, hardware (e.g., circuitry), firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present application are all or partially produced. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
In the present application, "at least one" means one or more, and "a plurality" means two or more. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b, or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or plural.
It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. A photovoltaic power prediction method based on illumination images, characterized in that it is applied to a server, the method comprising:
collecting a plurality of outdoor illumination images corresponding to a photovoltaic system according to a periodic collection strategy, and screening the collected outdoor illumination images to obtain a target illumination image, wherein the periodic collection strategy is set based on periodic changes of illumination intensity and comprises circadian periodic changes and seasonal periodic changes, the periodic collection strategy comprises a collection sub-strategy, and the collection sub-strategy comprises at least three times of outdoor illumination image collection;
inputting the target illumination image into a trained illumination intensity detection model to obtain illumination intensity estimation;
inputting the target illumination image into a trained illumination intensity detection model, wherein the method comprises the steps of inputting the target illumination image and a V-channel numerical value into the trained illumination intensity detection model;
the illumination intensity detection model is a neural network structure model and comprises an input layer, a convolution layer and a 9-layer residual pouring module connected with the convolution layer, wherein the residual pouring module is connected with two full-connection layers, and the two full-connection layers are respectively configured with an activation function and batch normalization operation and respectively output the probability distribution of the sun direction and the atmospheric turbidity;
The method for obtaining the illumination intensity estimation comprises the steps of:
inputting the target illumination image into a trained illumination intensity detection model, and outputting probability distribution and atmospheric turbidity of the sun direction corresponding to the target illumination image by the illumination intensity detection model;
obtaining illumination intensity estimation based on the probability distribution of the solar direction, the atmospheric turbidity and a solar illumination constant, and determining the illumination intensity estimation;
and obtaining a power coefficient based on the illumination intensity estimation and the standard illumination intensity, and obtaining a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
2. The photovoltaic power prediction method based on illumination images according to claim 1, wherein collecting a plurality of outdoor illumination images corresponding to a photovoltaic system according to a periodic collection strategy comprises: and acquiring outdoor illumination images in the corresponding season sections according to the seasonal variation, and acquiring a plurality of outdoor illumination images in the corresponding unit time sections according to the diurnal variation.
3. The illumination image-based photovoltaic power generation method according to claim 2, wherein acquiring a plurality of outdoor illumination images for a corresponding unit time period in accordance with a circadian variation comprises: and acquiring at least three outdoor illumination images in a corresponding unit time period according to the circadian cycle variation.
4. The photovoltaic power prediction method based on illumination images according to claim 3, wherein the step of screening the collected plurality of outdoor illumination images to obtain a target illumination image comprises the steps of: acquiring the gray scale of at least three outdoor illumination images, acquiring at least three gray scale ratio values among the at least three outdoor illumination images, and taking the finally acquired outdoor illumination images as target illumination images when the gray scale ratio values are within a preset threshold range; and when the gray ratio is not in the preset threshold range, eliminating the outdoor illumination image corresponding to the ratio exceeding the preset threshold range, and taking the last acquired image in the rest outdoor illumination images as a target illumination image, wherein the target illumination image is an RGB image.
5. The method of claim 4, further comprising performing color space conversion processing on the target illumination image before inputting the target illumination image to the trained illumination intensity detection model, converting the target illumination image into HSV color space, and outputting three channels respectively, wherein the value of the H channel is the hue of the color, the value of the S channel is the saturation of the color, and the value of the V channel is the brightness value of the color.
6. The method of claim 5, further comprising extracting a V-channel value before inputting the target illumination image to the trained illumination intensity detection model.
7. The method of claim 1, wherein the solar illumination constant is 2.838 x 10 27 cd。
8. A photovoltaic power generation apparatus, the apparatus comprising:
the outdoor illumination image acquisition module is used for acquiring a plurality of outdoor illumination images corresponding to the photovoltaic system according to a periodic acquisition strategy and screening the acquired outdoor illumination images to obtain a target illumination image;
the illumination intensity estimation module is used for inputting the target illumination image into the trained illumination intensity detection model to obtain illumination intensity estimation;
inputting the target illumination image into a trained illumination intensity detection model, wherein the method comprises the steps of inputting the target illumination image and a V-channel numerical value into the trained illumination intensity detection model;
the illumination intensity detection model is a neural network structure model and comprises an input layer, a convolution layer and a 9-layer residual pouring module connected with the convolution layer, wherein the residual pouring module is connected with two full-connection layers, and the two full-connection layers are respectively configured with an activation function and batch normalization operation and respectively output the probability distribution of the sun direction and the atmospheric turbidity;
The method for obtaining the illumination intensity estimation comprises the steps of:
inputting the target illumination image into a trained illumination intensity detection model, and outputting probability distribution and atmospheric turbidity of the sun direction corresponding to the target illumination image by the illumination intensity detection model;
obtaining illumination intensity estimation based on the probability distribution of the solar direction, the atmospheric turbidity and a solar illumination constant, and determining the illumination intensity estimation;
and the photovoltaic power prediction module is used for obtaining a power coefficient based on the illumination intensity estimation and the standard illumination intensity and obtaining a photovoltaic power prediction result based on the power coefficient and the maximum power of the photovoltaic system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310708777.2A CN116957223B (en) | 2023-06-14 | 2023-06-14 | Photovoltaic power prediction method and device based on illumination image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310708777.2A CN116957223B (en) | 2023-06-14 | 2023-06-14 | Photovoltaic power prediction method and device based on illumination image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116957223A CN116957223A (en) | 2023-10-27 |
| CN116957223B true CN116957223B (en) | 2024-03-08 |
Family
ID=88446929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310708777.2A Active CN116957223B (en) | 2023-06-14 | 2023-06-14 | Photovoltaic power prediction method and device based on illumination image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116957223B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111402327A (en) * | 2020-03-17 | 2020-07-10 | 韶鼎人工智能科技有限公司 | Outdoor photo sun position estimation method based on full convolution neural network |
| CN111582555A (en) * | 2020-04-19 | 2020-08-25 | 天津大学 | Photovoltaic power prediction method based on foundation cloud picture image characteristics |
| CN111815038A (en) * | 2020-06-24 | 2020-10-23 | 山东大学 | Photovoltaic ultra-short term prediction method and system |
| CN113159466A (en) * | 2021-05-27 | 2021-07-23 | 沃太能源股份有限公司 | Short-time photovoltaic power generation prediction system and method |
| CN113496311A (en) * | 2021-06-25 | 2021-10-12 | 国网山东省电力公司济宁供电公司 | Photovoltaic power station generated power prediction method and system |
| CN113537193A (en) * | 2021-07-15 | 2021-10-22 | Oppo广东移动通信有限公司 | Illumination estimation method, illumination estimation device, storage medium, and electronic apparatus |
| CN113762603A (en) * | 2021-08-13 | 2021-12-07 | 广西大学 | Photovoltaic base station short-term photovoltaic power prediction method based on improved sparrow algorithm optimization |
| CN114971062A (en) * | 2022-06-14 | 2022-08-30 | 中国华能集团清洁能源技术研究院有限公司 | Photovoltaic power prediction method and device |
| CN115936180A (en) * | 2022-11-02 | 2023-04-07 | 国网北京市电力公司 | Photovoltaic power generation power prediction method and device and computer equipment |
| CN116029440A (en) * | 2023-01-18 | 2023-04-28 | 阳光电源（上海）有限公司 | Ultra-short-term power prediction method and device for photovoltaic power station |
| CN116247654A (en) * | 2022-12-30 | 2023-06-09 | 山西省能源互联网研究院 | Photovoltaic power generation prediction method based on differentiable neural computer |
-
2023
- 2023-06-14 CN CN202310708777.2A patent/CN116957223B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111402327A (en) * | 2020-03-17 | 2020-07-10 | 韶鼎人工智能科技有限公司 | Outdoor photo sun position estimation method based on full convolution neural network |
| CN111582555A (en) * | 2020-04-19 | 2020-08-25 | 天津大学 | Photovoltaic power prediction method based on foundation cloud picture image characteristics |
| CN111815038A (en) * | 2020-06-24 | 2020-10-23 | 山东大学 | Photovoltaic ultra-short term prediction method and system |
| CN113159466A (en) * | 2021-05-27 | 2021-07-23 | 沃太能源股份有限公司 | Short-time photovoltaic power generation prediction system and method |
| CN113496311A (en) * | 2021-06-25 | 2021-10-12 | 国网山东省电力公司济宁供电公司 | Photovoltaic power station generated power prediction method and system |
| CN113537193A (en) * | 2021-07-15 | 2021-10-22 | Oppo广东移动通信有限公司 | Illumination estimation method, illumination estimation device, storage medium, and electronic apparatus |
| CN113762603A (en) * | 2021-08-13 | 2021-12-07 | 广西大学 | Photovoltaic base station short-term photovoltaic power prediction method based on improved sparrow algorithm optimization |
| CN114971062A (en) * | 2022-06-14 | 2022-08-30 | 中国华能集团清洁能源技术研究院有限公司 | Photovoltaic power prediction method and device |
| CN115936180A (en) * | 2022-11-02 | 2023-04-07 | 国网北京市电力公司 | Photovoltaic power generation power prediction method and device and computer equipment |
| CN116247654A (en) * | 2022-12-30 | 2023-06-09 | 山西省能源互联网研究院 | Photovoltaic power generation prediction method based on differentiable neural computer |
| CN116029440A (en) * | 2023-01-18 | 2023-04-28 | 阳光电源（上海）有限公司 | Ultra-short-term power prediction method and device for photovoltaic power station |
Non-Patent Citations (2)
| Title |
|---|
| A simplified methodology to estimate solar irradiance and atmospheric turbidity from ambient temperature and relative humidity;O. Behar 等;《Renewable and Sustainable Energy Reviews》(第116期);1-3 * |
| 基于地基云图的光伏发电系统超短期功率预测;韩阳;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》(第01期);C042-1689 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116957223A (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zeng et al. | A reversal in global terrestrial stilling and its implications for wind energy production | |
| Kabilan et al. | Short‐Term Power Prediction of Building Integrated Photovoltaic (BIPV) System Based on Machine Learning Algorithms | |
| Wang et al. | Comparative study of machine learning approaches for predicting short-term photovoltaic power output based on weather type classification | |
| CN110909919A (en) | Photovoltaic power prediction method of depth neural network model with attention mechanism fused | |
| CN103390199A (en) | Photovoltaic power generation capacity/power prediction device | |
| CN110929953A (en) | Photovoltaic power station ultra-short term output prediction method based on cluster analysis | |
| Michael et al. | Short-term solar irradiance forecasting based on a novel Bayesian optimized deep Long Short-Term Memory neural network | |
| Chen et al. | Short-term load forecasting based on deep learning for end-user transformer subject to volatile electric heating loads | |
| Aler et al. | A study of machine learning techniques for daily solar energy forecasting using numerical weather models | |
| CN114707688A (en) | Photovoltaic power ultra-short-term prediction method based on satellite cloud chart and space-time neural network | |
| Iyengar et al. | A cloud-based black-box solar predictor for smart homes | |
| CN112653198A (en) | Wind power output scene generation method and system based on prediction box | |
| Hu et al. | Short-term photovoltaic power prediction based on similar days and improved SOA-DBN model | |
| Huang et al. | Ultra‐short‐term photovoltaic power forecasting of multifeature based on hybrid deep learning | |
| CN116128170A (en) | Photovoltaic power station power ultra-short-term prediction method and device and related equipment | |
| Bandong et al. | Performance ratio estimation and prediction of solar power plants using machine learning to improve energy reliability | |
| Dwivedi et al. | Climate change: Prediction of solar radiation using advanced machine learning techniques | |
| CN116722544B (en) | Distributed photovoltaic short-term prediction method and device, electronic equipment and storage medium | |
| CN117613883A (en) | Method and device for predicting generated power, computer equipment and storage medium | |
| CN113240217A (en) | Photovoltaic power generation prediction method and device based on integrated prediction model | |
| Zamanidou et al. | Day‐ahead scheduling of a hybrid renewable energy system based on generation forecasting using a deep‐learning approach | |
| CN116957223B (en) | Photovoltaic power prediction method and device based on illumination image | |
| Nijhuis et al. | Prediction of power fluctuation classes for photovoltaic installations and potential benefits of dynamic reserve allocation | |
| CN111177278A (en) | Grid user short-term load prediction real-time processing tool | |
| Succetti et al. | Multivariate prediction of energy time series by autoencoded LSTM networks |
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 |