CN115759483B - Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium - Google Patents
Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115759483B CN115759483B CN202310015601.9A CN202310015601A CN115759483B CN 115759483 B CN115759483 B CN 115759483B CN 202310015601 A CN202310015601 A CN 202310015601A CN 115759483 B CN115759483 B CN 115759483B
- Authority
- CN
- China
- Prior art keywords
- prediction
- field
- irradiance
- solar irradiance
- data
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- 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
Abstract
The invention relates to the technical field of photovoltaic power generation, and provides a solar irradiance prediction method of a photovoltaic electric field, electronic equipment and a storage medium, wherein the method comprises the following steps: constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode; determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation; and inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value corresponding to the prediction time period, which is output by the irradiance prediction model. The invention completes the accurate prediction of solar irradiance from the angle of acquiring more practical characteristic data.
Description
Technical Field
The invention relates to the technical field of photovoltaic power generation, in particular to a solar irradiance prediction method for a photovoltaic electric field, electronic equipment and a storage medium.
Background
The output of a photovoltaic system is basically determined by the incident solar irradiance, so irradiance prediction is the most important step in most photovoltaic power prediction methods, and accurate solar irradiance prediction is an important issue.
Irradiance prediction methods can be categorized according to the input data used, which also determines the prediction horizon. The time sequence based on online irradiance measurement can be applied to a very short time scale, cloud cover information of a ground sky imager can be used for calculating predictions of high time and spatial resolution for a time range within an hour, satellite data has a good irradiance prediction effect for 6 hours in advance, and for longer prediction, a prediction result based on numerical weather prediction is the most commonly used choice.
Cloud is one of the most important factors determining the incident intensity of surface solar radiation. The formation of mid-latitude clouds is generally affected by weather-scale movements, and the formation of lamellar clouds is generally due to mid-latitude cyclones with elevation of the front. Under the influence of weather conditions that the rising movement of air flow is strong and the atmospheric water vapor content is high, rain clouds can appear in summer, and sometimes clear sky clouds exist. The shallow convection cloud space scale is only a few hundred meters to 1-2 km, the life time is 10-50 minutes, and the time space is changed greatly, so that the rapid change of local solar irradiance is caused. In addition, under the unstable condition of the atmospheric junction, strong convection events with small scale and time period are easily formed by the ascending motion, so that the local cloud amount is changed, and the rapid change of solar irradiance is also caused. The grid scale of numerical forecasting is generally several kilometers, cloud generation and movement with smaller space and time scale are difficult to capture, and the solar irradiance change forecasting difficulty for high time-space resolution is high.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a solar irradiance prediction method for a photovoltaic electric field, electronic equipment and a storage medium.
In a first aspect, the present invention provides a method for predicting solar irradiance of a photovoltaic electric field, comprising:
constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode;
determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation;
inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value corresponding to the prediction time period output by the irradiance prediction model;
the irradiance prediction model is obtained by taking solar irradiance according to space potential field characteristic data in sample data and the sample data as input through machine learning training and is used for predicting solar irradiance.
In one embodiment, the method further comprises the step of obtaining an irradiance prediction model, comprising:
acquiring n pieces of historical data positioned in the space-shaped potential field in a preset time period; each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by sequentially arranging potential heights, temperatures and humidities on grid points in the space potential field;
constructing an m multiplied by n feature matrix according to n pieces of historical data;
obtaining a prediction feature vector obtained based on n pieces of historical data prediction;
determining the similarity between each piece of historical data and the prediction feature matrix according to the m multiplied by n feature matrix and the prediction feature vector;
determining w pieces of historical data from the n pieces of historical data as sample data according to the similarity;
and training according to the feature matrix and solar irradiance in the sample data to obtain an irradiance prediction model.
In one embodiment, determining the similarity between each piece of historical data and the predicted feature vector from the m×n feature matrix and the predicted feature vector includes:
calculating absolute difference vectors between corresponding feature vectors and predicted feature vectors of each piece of historical data in the m multiplied by n feature matrix one by one;
normalizing the absolute difference vector to determine a normalized absolute difference vector;
the values within the absolute difference vectors are summed to determine the similarity between each piece of historical data and the predicted feature vector.
In one embodiment, obtaining an irradiance prediction model according to the feature matrix and solar irradiance training in the sample data comprises:
training an initial model based on sample data, and performing a first step on the modelkDuring iterative training, predicting a feature matrix in sample data by adopting a model in the current training, determining solar irradiance, and taking the solar irradiance and the feature matrix obtained by prediction as new sample data;
and continuing training the model in the current training based on the new sample data, and updating model parameters when the iteration times meet the preset number, so as to determine an irradiance prediction model.
In a second aspect, the present invention provides a photovoltaic electric field solar irradiance prediction apparatus, comprising:
the construction module is used for constructing a space shape potential field taking the photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and the space shape potential field is a multi-level grid space simulated by adopting a mesoscale numerical mode;
the determining module is used for determining the spatial shape potential field characteristic data in the prediction time period by adopting mesoscale numerical mode simulation;
the prediction module is used for inputting the space situation field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value which is output by the irradiance prediction model and corresponds to the prediction time period;
the irradiance prediction model is obtained by taking solar irradiance according to space potential field characteristic data in sample data and the sample data as input through machine learning training and is used for predicting solar irradiance.
In one embodiment, the apparatus further comprises a generation module for:
acquiring n pieces of historical data positioned in the space-shaped potential field in a preset time period; each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by sequentially arranging potential heights, temperatures and humidities on grid points in the space potential field;
constructing an m multiplied by n feature matrix according to n pieces of historical data;
obtaining a prediction feature vector obtained based on n pieces of historical data prediction;
determining the similarity between each piece of historical data and the prediction feature matrix according to the m multiplied by n feature matrix and the prediction feature vector;
determining w pieces of historical data from the n pieces of historical data as sample data according to the similarity;
and training according to the feature matrix and solar irradiance in the sample data to obtain an irradiance prediction model.
In one embodiment, the generating module is specifically configured to, in a process of determining a similarity between each piece of history data and the predicted feature vector according to the mxn feature matrix and the predicted feature vector:
calculating absolute difference vectors between corresponding feature vectors and predicted feature vectors of each piece of historical data in the m multiplied by n feature matrix one by one;
normalizing the absolute difference vector to determine a normalized absolute difference vector;
the values within the absolute difference vectors are summed to determine the similarity between each piece of historical data and the predicted feature vector.
In one embodiment, the generating module is specifically configured to, in a process of obtaining an irradiance prediction model according to the feature matrix in the sample data and solar irradiance training:
training an initial model based on sample data, and performing a first step on the modelkDuring iterative training, predicting a feature matrix in sample data by adopting a model in the current training, determining solar irradiance, and taking the solar irradiance and the feature matrix obtained by prediction as new sample data;
and continuing training the model in the current training based on the new sample data, and updating model parameters when the iteration times meet the preset number, so as to determine an irradiance prediction model.
In a third aspect, the present invention provides an electronic device comprising a memory and a memory storing a computer program, the processor implementing the steps of the photovoltaic electric field solar irradiance prediction method of the first aspect when executing the program.
In a fourth aspect, the present invention provides a processor-readable storage medium storing a computer program for causing the processor to perform the steps of the photovoltaic electric field solar irradiance prediction method of the first aspect.
According to the solar irradiance prediction method, the electronic equipment and the storage medium for the photovoltaic electric field, provided by the invention, the spatial shape potential field characteristic data in the prediction time period is determined by adopting the mesoscale numerical mode simulation, the spatial shape potential field characteristic data is input into the irradiance prediction model, the solar irradiance prediction value corresponding to the prediction time period, which is output by the irradiance prediction model, is obtained, and the accurate prediction of the solar irradiance is completed from the aspect of obtaining more practical characteristic data.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a photovoltaic electric field solar irradiance prediction method provided by the invention;
FIG. 2 is a schematic structural view of a photovoltaic electric field solar irradiance prediction apparatus provided by the invention;
fig. 3 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. 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.
Fig. 1 shows a schematic flow chart of a photovoltaic electric field solar irradiance prediction method provided by the invention, and referring to fig. 1, the method comprises:
11. constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode;
12. determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation;
13. inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value which is output by the irradiance prediction model and corresponds to a prediction time period;
the irradiance prediction model is obtained by taking solar irradiance according to space potential field characteristic data in sample data and sample data as input and training through machine learning and is used for predicting the solar irradiance.
For the steps 11 to 13, it should be noted that, in the present invention, a multi-level grid space is simulated by using a mesoscale numerical mode. The spatial situational field has a potential height field, a temperature field and a humidity field.
The position of the photovoltaic electric field is taken as a central coordinate, a certain grid range is selected in the horizontal direction, and a certain layer number (for example, 850hPa, 700hPa and 500hPa of air pressure layers) is selected in the vertical direction, so that a three-dimensional space is formed and is used as a space shape potential field of the electric field.
In the invention, historical data in a space shape potential field in a first preset time period is obtained, wherein the historical data comprises potential height, temperature and humidity on each grid point in the space shape potential field, and the potential height, the temperature and the humidity are meteorological factors on each grid point.
The selection mode of the historical data is as follows:
and selecting the data of the last 15 days of the current moment and 15 days before and after the history of the last h calendar, wherein the time resolution of the data is the same as the resolution of the actually measured solar irradiance of the station, and is generally 15 minutes.
In the invention, the spatial shape potential field characteristic data in the prediction time period is determined by adopting mesoscale numerical mode simulation. The spatial situation field characteristic data in the predicted time period determined by the mesoscale numerical mode simulation is closer to the actual situation in the predicted time period, and therefore, the predicted spatial situation field characteristic data is used as input data of an irradiance prediction model, and then the irradiance prediction model outputs a solar irradiance predicted value corresponding to the predicted time period.
In the invention, the irradiance prediction model is a model which is obtained by taking solar irradiance according to spatial potential field characteristic data in sample data and sample data as input and through machine learning training and is used for predicting solar irradiance.
According to the solar irradiance prediction method for the photovoltaic electric field, provided by the invention, the spatial situation field characteristic data in the prediction time period is determined by adopting the mesoscale numerical mode simulation, the spatial situation field characteristic data is input into the irradiance prediction model, the solar irradiance prediction value which is output by the irradiance prediction model and corresponds to the prediction time period is obtained, and the accurate prediction of the solar irradiance is completed from the aspect of obtaining more practical characteristic data.
In the further method, the process of obtaining the irradiance prediction model is mainly explained, specifically:
acquiring n pieces of historical data positioned in the space-shaped potential field in a second preset time period; each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by sequentially arranging potential heights, temperatures and humidities on grid points in the space potential field;
constructing an m multiplied by n feature matrix according to n pieces of historical data;
obtaining a prediction feature vector obtained based on n pieces of historical data prediction;
determining the similarity between each piece of historical data and the prediction feature matrix according to the m multiplied by n feature matrix and the prediction feature vector;
determining w pieces of historical data from the n pieces of historical data according to the similarity as sample data;
and training according to the feature matrix and solar irradiance in the sample data to obtain an irradiance prediction model.
In this regard, in the present invention, n pieces of history data located in the space-shaped potential field during a preset period of time are acquired, where the history data includes potential heights, temperatures, and humidities at each grid point in the space-shaped potential field, and the potential heights, temperatures, and humidities are weather factors at each grid point.
The selection mode of the historical data is as follows:
and selecting the data of the last 15 days of the current moment and 15 days before and after the history of the last h calendar, wherein the time resolution of the data is the same as the resolution of the actually measured solar irradiance of the station, and is generally 15 minutes.
Each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by arranging potential heights, temperatures and humidities on grid points in a space potential field in sequence. Whereby an mxn feature matrix is constructed based on the n pieces of history data.
The feature matrix is as follows:
acquisition is based onnThe predicted feature vector predicted by the historical data is as follows:
and determining the similarity between each piece of historical data and the prediction feature matrix according to the m multiplied by n feature matrix and the prediction feature vector.
Namely: absolute difference vectors between corresponding feature vectors and predicted feature vectors of each piece of historical data in the m×n feature matrix are calculated one by one as follows:
and carrying out normalization processing on the absolute difference vector to determine a normalized absolute difference vector. The following are provided:
then, the values in the absolute difference vectors are summed to determine the similarity between each piece of history data and the predicted feature vector as follows:
In the present invention, the historical data with the similarity of the first 40% may be selected as the sample data.
In the further method, the processing procedure of obtaining the irradiance prediction model according to the feature matrix in the sample data and solar irradiance training is mainly explained, and the method specifically comprises the following steps:
training an initial model based on sample data, and performing a first step on the modelkWhen iterative training is performed for the second time, predicting a feature matrix in sample data by adopting a model in the current training, determining solar irradiance, and taking the feature matrix obtained by prediction as new sample data;
and continuing training the model in the current training based on the new sample data, and updating model parameters when the iteration times meet the preset number, so as to determine an irradiance prediction model.
In this regard, it should be noted that the iterative training method provided by the present invention predicts the sample data of the previous k-1 using the current model, and obtains the model in the current training after the initial model is trained. And then taking the predicted solar irradiance and the feature matrix as new sample data, and adding the new sample data into subsequent training.
After the training data of the kth iteration is summarized, the size of the training set is k times of the initial size, the model is continuously trained on the new training set before the next iteration, the gradient descent steps are kept the same, and when the iteration times meet the preset number, model parameters are updated to determine a path determination model. The model trained in this way enables a more accurate solar irradiance to be determined.
The photovoltaic electric field solar irradiance prediction apparatus provided by the invention is described below, and the photovoltaic electric field solar irradiance prediction apparatus described below and the photovoltaic electric field solar irradiance prediction method described above can be referred to correspondingly.
Fig. 2 shows a schematic flow chart of a photovoltaic electric field solar irradiance prediction apparatus provided by the present invention, referring to fig. 2, the apparatus includes a construction module 21, an acquisition module 22, a determination module 23, and a prediction module 24, wherein:
a construction module 21, configured to construct a space shape potential field with a photovoltaic electric field as a central coordinate, where the space shape potential field includes a potential height field, a temperature field, and a humidity field, and the space shape potential field is a multi-level grid space simulated by adopting a mesoscale numerical mode;
a determining module 23, configured to determine spatial shape potential field characteristic data in a predicted time period by adopting mesoscale numerical mode simulation;
a prediction module 24, configured to input the spatial situational field characteristic data into an irradiance prediction model, and obtain a solar irradiance prediction value corresponding to the predicted time period output by the irradiance prediction model;
the irradiance prediction model is obtained by taking solar irradiance according to space potential field characteristic data in sample data and the sample data as input through machine learning training and is used for predicting solar irradiance.
In a further arrangement of the above arrangement, the arrangement further comprises a generation module for:
acquiring n pieces of historical data positioned in the space-shaped potential field in a second preset time period; each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by sequentially arranging potential heights, temperatures and humidities on grid points in the space potential field;
constructing an m multiplied by n feature matrix according to n pieces of historical data;
obtaining a prediction feature vector obtained based on n pieces of historical data prediction;
determining the similarity between each piece of historical data and the prediction feature matrix according to the m multiplied by n feature matrix and the prediction feature vector;
determining w pieces of historical data from the n pieces of historical data as sample data according to the similarity;
and training according to the feature matrix and solar irradiance in the sample data to obtain an irradiance prediction model.
In a further apparatus of the above apparatus, the generating module is specifically configured to, in a process of determining a similarity between each piece of history data and the predicted feature vector based on the m×n feature matrix and the predicted feature vector:
calculating absolute difference vectors between corresponding feature vectors and predicted feature vectors of each piece of historical data in the m multiplied by n feature matrix one by one;
normalizing the absolute difference vector to determine a normalized absolute difference vector;
the values within the absolute difference vectors are summed to determine the similarity between each piece of historical data and the predicted feature vector.
In a further apparatus of the above apparatus, the generating module is specifically configured to:
training an initial model based on sample data, and performing a first step on the modelkDuring iterative training, predicting a feature matrix in sample data by adopting a model in the current training, determining solar irradiance, and taking the solar irradiance and the feature matrix obtained by prediction as new sample data;
and continuing training the model in the current training based on the new sample data, and updating model parameters when the iteration times meet the preset number, so as to determine an irradiance prediction model.
Since the apparatus according to the embodiment of the present invention is the same as the method according to the above embodiment, the details of the explanation will not be repeated here.
It should be noted that, in the embodiment of the present invention, the related functional modules may be implemented by a hardware processor (hardware processor).
According to the solar irradiance prediction device for the photovoltaic electric field, provided by the invention, the spatial situation field characteristic data in the prediction time period is determined by adopting the mesoscale numerical mode simulation, the spatial situation field characteristic data is input into the irradiance prediction model, the solar irradiance prediction value which is output by the irradiance prediction model and corresponds to the prediction time period is obtained, and the accurate prediction of the solar irradiance is completed from the aspect of obtaining more practical characteristic data.
Fig. 3 illustrates a physical schematic diagram of an electronic device, as shown in fig. 3, where the electronic device may include: a processor (processor) 31, a communication interface (Communication Interface) 32, a memory (memory) 33 and a communication bus 34, wherein the processor 31, the communication interface 32 and the memory 33 communicate with each other through the communication bus 34. The processor 31 may invoke a computer program in the memory 33 to perform the steps of the photovoltaic farm solar irradiance prediction method, including, for example: constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode; determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation; and inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value corresponding to the prediction time period, which is output by the irradiance prediction model.
Further, the logic instructions in the memory 33 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention 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, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RAM, randomAccessMemory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, are capable of performing the steps of a photovoltaic farm solar irradiance prediction method, for example comprising: constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode; determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation; and inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value corresponding to the prediction time period, which is output by the irradiance prediction model.
In another aspect, embodiments of the present invention further provide a processor-readable storage medium storing a computer program for causing the processor to perform the steps of a photovoltaic electric field solar irradiance prediction method, for example, including: constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode; determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation; and inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value corresponding to the prediction time period, which is output by the irradiance prediction model.
The processor-readable storage medium may be any available medium or data storage device that can be accessed by a processor, including, but not limited to, magnetic storage (e.g., floppy disks, hard disks, magnetic tape, magneto-optical disks (MOs), etc.), optical storage (e.g., CD, DVD, BD, HVD, etc.), semiconductor storage (e.g., ROM, EPROM, EEPROM, nonvolatile storage (NAND FLASH), solid State Disk (SSD)), and the like.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A photovoltaic electric field solar irradiance prediction method, comprising:
constructing a space shape potential field taking a photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and is a multi-level grid space simulated by adopting a mesoscale numerical mode;
determining the space shape potential field characteristic data in a prediction time period by adopting mesoscale numerical mode simulation;
inputting the space situational field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value corresponding to the prediction time period output by the irradiance prediction model;
the irradiance prediction model is obtained by taking solar irradiance according to space shape potential field characteristic data in sample data and the sample data as input through machine learning training and is used for predicting solar irradiance;
the method further comprises the step of obtaining an irradiance prediction model, comprising:
acquiring n pieces of historical data positioned in the space-shaped potential field in a preset time period; each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by sequentially arranging potential heights, temperatures and humidities on grid points in the space potential field;
constructing an m multiplied by n feature matrix according to n pieces of historical data;
obtaining a prediction feature vector obtained based on n pieces of historical data prediction;
determining the similarity between each piece of historical data and the predictive feature vector according to the m multiplied by n feature matrix and the predictive feature vector;
determining w pieces of historical data from the n pieces of historical data as sample data according to the similarity;
and training according to the feature matrix and solar irradiance in the sample data to obtain an irradiance prediction model.
2. The method of claim 1, wherein determining the similarity between each piece of historical data and the predicted feature vector from the m x n feature matrix and the predicted feature vector comprises:
calculating absolute difference vectors between corresponding feature vectors and predicted feature vectors of each piece of historical data in the m multiplied by n feature matrix one by one;
normalizing the absolute difference vector to determine a normalized absolute difference vector;
the values within the absolute difference vectors are summed to determine the similarity between each piece of historical data and the predicted feature vector.
3. The method according to claim 1, wherein the training to obtain an irradiance prediction model according to the feature matrix and solar irradiance in the sample data comprises:
training an initial model based on sample data, and performing a first step on the modelkDuring iterative training, predicting a feature matrix in sample data by adopting a model in the current training, determining solar irradiance, and taking the solar irradiance and the feature matrix obtained by prediction as new sample data;
and continuing training the model in the current training based on the new sample data, and updating model parameters when the iteration times meet the preset number, so as to determine an irradiance prediction model.
4. A photovoltaic electric field solar irradiance prediction apparatus, comprising:
the construction module is used for constructing a space shape potential field taking the photovoltaic electric field as a central coordinate, wherein the space shape potential field is provided with a potential height field, a temperature field and a humidity field, and the space shape potential field is a multi-level grid space simulated by adopting a mesoscale numerical mode;
the determining module is used for determining the spatial shape potential field characteristic data in the prediction time period by adopting mesoscale numerical mode simulation;
the prediction module is used for inputting the space situation field characteristic data into an irradiance prediction model to obtain a solar irradiance prediction value which is output by the irradiance prediction model and corresponds to the prediction time period;
the irradiance prediction model is obtained by taking solar irradiance according to space shape potential field characteristic data in sample data and the sample data as input through machine learning training and is used for predicting solar irradiance;
the apparatus further comprises a generation module for:
acquiring n pieces of historical data positioned in the space-shaped potential field in a preset time period; each piece of historical data comprises m meteorological factors, wherein the m meteorological factors are formed by sequentially arranging potential heights, temperatures and humidities on grid points in the space potential field;
constructing an m multiplied by n feature matrix according to n pieces of historical data;
obtaining a prediction feature vector obtained based on n pieces of historical data prediction;
determining the similarity between each piece of historical data and the predictive feature vector according to the m multiplied by n feature matrix and the predictive feature vector;
determining w pieces of historical data from the n pieces of historical data as sample data according to the similarity;
and training according to the feature matrix and solar irradiance in the sample data to obtain an irradiance prediction model.
5. The photovoltaic electric field solar irradiance prediction apparatus of claim 4, wherein the generating module is configured to, in a process of determining a similarity between each piece of history data and the predicted feature vector according to the m x n feature matrix and the predicted feature vector:
calculating absolute difference vectors between corresponding feature vectors and predicted feature vectors of each piece of historical data in the m multiplied by n feature matrix one by one;
normalizing the absolute difference vector to determine a normalized absolute difference vector;
the values within the absolute difference vectors are summed to determine the similarity between each piece of historical data and the predicted feature vector.
6. The photovoltaic electric field solar irradiance prediction apparatus of claim 4, wherein the generating module is configured to, in a process of obtaining an irradiance prediction model according to the feature matrix and solar irradiance training in the sample data:
training an initial model based on sample data, and performing a first step on the modelkDuring iterative training, predicting a feature matrix in sample data by adopting a model in the current training, determining solar irradiance, and taking the solar irradiance and the feature matrix obtained by prediction as new sample data;
and continuing training the model in the current training based on the new sample data, and updating model parameters when the iteration times meet the preset number, so as to determine an irradiance prediction model.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the steps of the photovoltaic electric field solar irradiance prediction method of any of claims 1 to 3.
8. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the photovoltaic electric field solar irradiance prediction method of any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310015601.9A CN115759483B (en) | 2023-01-06 | 2023-01-06 | Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310015601.9A CN115759483B (en) | 2023-01-06 | 2023-01-06 | Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115759483A CN115759483A (en) | 2023-03-07 |
| CN115759483B true CN115759483B (en) | 2023-06-20 |
Family
ID=85348256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310015601.9A Active CN115759483B (en) | 2023-01-06 | 2023-01-06 | Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115759483B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10989839B1 (en) * | 2017-08-29 | 2021-04-27 | University Of Hawai'i | Ground-based sky imaging and irradiance prediction system |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955749B (en) * | 2014-03-19 | 2018-03-09 | 内蒙古电力勘测设计院有限责任公司 | Direct solar radiation value Forecasting Methodology and system |
| CA2996216C (en) * | 2015-08-31 | 2023-09-26 | Green Power Labs Inc. | Method and system for solar power forecasting |
| CN105373857A (en) * | 2015-11-27 | 2016-03-02 | 许昌许继软件技术有限公司 | Photovoltaic power station irradiance prediction method |
| CN109829572B (en) * | 2019-01-14 | 2021-03-12 | 国网江苏省电力有限公司苏州供电分公司 | Photovoltaic power generation power prediction method under thunder and lightning weather |
| CN109636076A (en) * | 2019-02-14 | 2019-04-16 | 国能日新科技股份有限公司 | A kind of distributed photovoltaic power prediction main station system |
| CN114492915A (en) * | 2021-12-02 | 2022-05-13 | 中国电力科学研究院有限公司 | Method and device for predicting solar irradiance |
| CN114266416A (en) * | 2021-12-30 | 2022-04-01 | 南京洁源电力科技发展有限公司 | Photovoltaic power generation power short-term prediction method and device based on similar days and storage medium |
| CN114492181A (en) * | 2022-01-18 | 2022-05-13 | 汉谷云智(武汉)科技有限公司 | Photovoltaic ultra-short term power prediction method and device based on adaptive meteorological classification |
-
2023
- 2023-01-06 CN CN202310015601.9A patent/CN115759483B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10989839B1 (en) * | 2017-08-29 | 2021-04-27 | University Of Hawai'i | Ground-based sky imaging and irradiance prediction system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115759483A (en) | 2023-03-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5888640B2 (en) | Photovoltaic power generation prediction apparatus, solar power generation prediction method, and solar power generation prediction program | |
| Carneiro et al. | Review on photovoltaic power and solar resource forecasting: current status and trends | |
| JP6543028B2 (en) | Apparatus and method for diagnosing abnormality | |
| TWI745907B (en) | A power generation amount prediction device, a power generation amount prediction method and program | |
| JP2009225550A (en) | Power demand predicting method and system, and power generation prediction method | |
| CN106651007A (en) | Method and device for GRU-based medium and long-term prediction of irradiance of photovoltaic power station | |
| WO2021051035A1 (en) | Spatio-temporal probabilistic forecasting of wind power output | |
| CN111815027A (en) | Photovoltaic station generated power prediction method and system | |
| CN103971167B (en) | Predict the method and system of the generated output of photovoltaic power generation equipment | |
| KR20200057822A (en) | optimaized water cooling apparatus and method for solar photovoltatic power generation module using rain water | |
| CN115759483B (en) | Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium | |
| CN112307672A (en) | BP neural network short-term wind power prediction method based on cuckoo algorithm optimization | |
| CN107358059A (en) | Short-term photovoltaic energy Forecasting Methodology and device | |
| CN113610665B (en) | Wind power generation power prediction method based on multi-delay output echo state network | |
| CN113568067B (en) | Numerical weather forecasting method and device, computer storage medium and electronic equipment | |
| CN115618714A (en) | Solar radiation probability forecasting method and related equipment | |
| CN115859765A (en) | Method, device, equipment and storage medium for predicting city expansion | |
| CN116415700A (en) | Wind energy numerical forecasting method and device combining artificial intelligence | |
| Gerards et al. | Pv predictions made easy: Flexibility through simplicity | |
| CN115599598B (en) | Power load sensing data recovery method and device | |
| KR102388292B1 (en) | Radiation parameterization system for correcting uncertainty of radiation parameterization emulator and operating method thereof | |
| CN117039894B (en) | Photovoltaic power short-term prediction method and system based on improved dung beetle optimization algorithm | |
| JP7272158B2 (en) | Power generation output calculation device and power generation output calculation method | |
| CN116822658B (en) | Early warning method and device for wind-solar resources, computer equipment and storage medium | |
| CN117713204A (en) | New energy power generation scene generation method and system considering space-time correlation |
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 |