Multi-source data fused vertical plane hourly total radiation calculation method and related equipment

Abstract

The invention discloses a multi-source data fused vertical plane hourly total radiation calculation method and related equipment, and the method comprises the steps: firstly obtaining multi-source data of a target area, including horizontal plane solar radiation, ground meteorology, satellite inversion data and astronomical parameters; judging data conditions of a target area, wherein the data conditions are divided into two types with and without horizontal plane solar radiation data; selecting a corresponding input parameter combination from the multi-source data according to the data condition, selecting the horizontal plane solar radiation data and the ground meteorological data when the horizontal plane solar radiation data exists, and selecting satellite inversion and astronomical parameters when the horizontal plane solar radiation data does not exist; then training a predefined machine learning model by using the selected parameter combination to obtain a trained vertical plane hourly total radiation calculation model; and finally, inputting current multi-source data into the model to obtain a vertical plane hourly total radiation calculation value of the target area. The method aims at solving the problem that an existing method depends on experience judgment on input variable selection, so that the adaptability and generalization ability of a model are insufficient in a data condition change scene.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of building energy-saving and low-carbon design, and relates to a vertical surface hourly total radiation calculation method fusing multi-source data and related equipment. BACKGROUND

[0002] Photovoltaic and photothermal envelope structures and solar energy systems are the main solar energy utilization methods, and the implementation of these technologies relies on high-precision building facade radiation data. The solar radiation received by the building facade is obviously affected by meteorological conditions and has strong hourly fluctuation, and this instability brings problems to the efficient operation of photovoltaic and photothermal systems. Therefore, obtaining accurate hourly total radiation data of each orientation vertical surface is one of the key foundations for realizing efficient utilization of building solar energy.

[0003] Instrument observation is the most accurate method to obtain solar radiation data. However, the national-level business ground radiation observation station only provides horizontal surface solar radiation data, and lacks oblique surface radiation observation data. Establishing a vertical surface radiation model through related parameters is an effective method to supplement the lack of radiation observation data. The vertical surface radiation is composed of direct radiation, scattered radiation and ground reflection, among which the direct radiation and ground reflection can be calculated by the horizontal surface total radiation, the incident angle and the ground reflectivity. However, due to the anisotropic distribution of sky scattered radiation, the calculation of scattered radiation is more complex. Existing researches mostly use the isotropic model and the anisotropic model to calculate the scattered radiation. The isotropic model assumes that the scattered radiation is uniformly distributed, which is suitable for overcast days. The anisotropic model assumes that the sky scattering is composed of three parts, which can further express the anisotropy of scattered radiation but is complex in calculation and requires many input parameters. There are also researches that use machine learning models to establish a vertical surface hourly total radiation model to balance the calculation accuracy and simplicity of the model. However, the selection of input variables in this method often depends on empirical judgment, which leads to insufficient adaptability and generalization ability of the constructed model in the scenario of changing data conditions. This limitation is more prominent in areas lacking ground measured radiation data. How to construct a reliable and generalizable vertical surface radiation calculation model has become a technical problem to be solved. SUMMARY

[0004] In view of the problems in the prior art, the application provides a vertical surface hourly total radiation calculation method fusing multi-source data and related equipment, which aims to overcome the problem that the existing method depends on empirical judgment in the selection of input variables, leading to insufficient adaptability and generalization ability of the model in the scenario of changing data conditions.

[0005] To solve the above technical problems, the application is implemented through the following technical solutions: According to a first aspect of the application, a vertical surface hourly total radiation calculation method fusing multi-source data is provided, comprising: acquire multi-source data of a target region, the multi-source data comprising horizontal plane solar radiation data, ground meteorological data, satellite inversion data and astronomical parameters; determine a data condition of the target region based on the multi-source data, the data condition comprising a horizontal plane solar radiation data condition or a horizontal plane solar radiation data-free condition; select a corresponding input parameter combination from the multi-source data according to the data condition, wherein: when the data condition is the horizontal plane solar radiation data condition, the selected input parameter combination comprises the horizontal plane solar radiation data and the ground meteorological data; and when the data condition is the horizontal plane solar radiation data-free condition, the selected input parameter combination comprises the satellite inversion data and the astronomical parameters; train a predefined machine learning model using the selected input parameter combination to obtain a trained vertical plane hourly total radiation calculation model; input current multi-source data into the trained vertical plane hourly total radiation calculation model to obtain a vertical plane hourly total radiation calculation value of the target region.

[0006] In a possible implementation manner of the first aspect, the horizontal plane solar radiation data comprises horizontal plane hourly total radiation and horizontal plane hourly scattered radiation. The ground meteorological data comprises hourly relative humidity, hourly sunshine duration and hourly air temperature. The satellite inversion data comprises hourly cloud amount, hourly ozone, hourly aerosol optical depth and hourly precipitation.

[0007] In a possible implementation manner of the first aspect, the determination of the data condition of the target region specifically comprises: if the horizontal plane solar radiation data exists, determining that the data condition is the horizontal plane solar radiation data condition; if the horizontal plane solar radiation data does not exist, determining that the data condition is the horizontal plane solar radiation data-free condition.

[0008] In a possible implementation manner of the first aspect, the input parameter combination is selected for different building orientations respectively, and the orientations comprise due east, due south, due west and due north.

[0009] In a possible implementation manner of the first aspect, the predefined machine learning model is an XGBoost model.

[0010] In a possible implementation manner of the first aspect, when training the XGBoost model, a grid search optimization or a genetic algorithm is used for parameter tuning to optimize hyperparameters, and the hyperparameters comprise maximum tree depth, learning rate, sub-sample ratio, feature sampling ratio and number of trees.

[0011] In a possible implementation manner of the first aspect, after obtaining the calculated value of the vertical surface hourly total radiation of the target region, the method further includes: The calculated value of the vertical surface hourly total radiation is visualized, and the visualization includes scatter plot visualization and time series plot visualization.

[0012] In a possible implementation manner of the first aspect, the astronomical parameters include a solar elevation angle, a solar azimuth angle, a solar incident angle of each orientation, a declination angle, a time angle, and astronomical radiation.

[0013] According to a second aspect of the present application, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method for calculating vertical surface hourly total radiation by fusing multi-source data when executing the computer program.

[0014] According to a third aspect of the present application, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the method for calculating vertical surface hourly total radiation by fusing multi-source data.

[0015] Compared with the prior art, the present application has at least the following beneficial effects: The method for calculating vertical surface hourly total radiation by fusing multi-source data provided by the present application can adaptively select input parameter combinations according to the data conditions of a target region, such as whether there is horizontal plane solar radiation data, overcome the limitation of relying on experience to select input variables, make the model be able to flexibly cope with the changes of different data conditions, make the prediction results more stable and have good generalization performance. By comprehensively utilizing multi-source data, including horizontal plane solar radiation data, ground meteorological data, satellite inversion data and astronomical parameters, in the area lacking horizontal plane solar radiation data, satellite inversion data and other parameters can be used as input to construct a reliable vertical surface hourly total radiation calculation model, solving the prediction problem existing in such area. Based on the data conditions, the corresponding input parameter combination is matched, and a machine learning model is trained, which can more accurately capture the hourly change characteristics of the vertical surface solar radiation, so as to obtain accurate vertical surface hourly total radiation data of each orientation. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the specific embodiments of the present application, the drawings needed in the description of the specific embodiments will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0017] Figure 1 is a flowchart of a vertical plane hourly total radiation calculation method that fuses multi-source data according to an embodiment of the present application.

[0018] Figure 2 is a flowchart of a vertical plane hourly total radiation calculation method that fuses multi-source data according to an embodiment of the present application.

[0019] Figure 3 is a fitting plot of the calculated values and the measured values of the horizontal plane solar radiation data under the optimal input parameter combination for each orientation in a region according to an embodiment. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be described below in detail with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0021] As shown in Figure 1 , the present application provides a vertical plane hourly total radiation calculation method that fuses multi-source data. The method intelligently judges the data status of a target region, adaptively selects an input parameter combination, and uses a machine learning model for training and prediction, thereby effectively solving the problem of insufficient model adaptability and generalization ability caused by the selection of input variables relying on experience in the prior art. The present application is particularly suitable for regions lacking ground radiation data and can reliably obtain vertical plane hourly total radiation data for each orientation.

[0022] In a typical embodiment of the present application, the calculation method of the vertical plane hourly total radiation includes the following steps: S1, obtaining multi-source data of a target region, the multi-source data including horizontal plane solar radiation data, ground meteorological data, satellite inversion data, and astronomical parameters.

[0023] Specifically, the horizontal plane solar radiation data can include horizontal plane hourly total radiation and horizontal plane hourly scattered radiation; the ground meteorological data includes hourly sunshine duration, hourly air temperature, and hourly relative humidity; the satellite inversion data can include hourly cloud cover, hourly ozone, hourly aerosol optical depth, and hourly precipitable water; and the astronomical parameters can include solar elevation angle, solar azimuth angle, solar incident angle for each orientation, declination angle, hour angle, and astronomical radiation. It should be understood that these data can be obtained through public databases, weather stations, or satellite remote sensing platforms and preprocessed to ensure data quality.

[0024] Preferably, the multi-source data is pre-processed and quality controlled. The pre-processing includes time series matching, and the quality control includes but is not limited to screening to remove missing values, outliers and extreme values. Specifically, the multi-source data is pre-processed according to the following rules: The multi-source data is time-unified to ensure that the time series of the hourly total radiation data of each orientation and the horizontal plane solar radiation data are consistent. Further, the data is quality controlled to eliminate abnormal data, such as data in which the horizontal plane hourly total radiation is greater than the hourly astronomical solar radiation data, and the solar elevation angle is less than 7 degrees.

[0025] S2, based on the multi-source data, determining the data condition of the target region, the data condition including a horizontal plane solar radiation data condition or a horizontal plane solar radiation data condition.

[0026] Specifically, based on the multi-source data obtained in step S1, the data condition of the target region is evaluated. The judgment logic is as follows: if there is horizontal plane solar radiation data, it is determined to be a horizontal plane solar radiation data condition; if there is no horizontal plane solar radiation data, it is determined to be a horizontal plane solar radiation data condition. Illustratively, the judgment process can be realized through an automated script or a data management tool to ensure rapid and accurate identification of the data condition.

[0027] S3, according to the data condition, selecting a corresponding input parameter combination from the multi-source data, the specific selection rules being: When the data condition is a horizontal plane solar radiation data condition, the selected input parameter combination includes the horizontal plane solar radiation data and the ground meteorological data; When the data condition is a horizontal plane solar radiation data condition, the selected input parameter combination includes the satellite inversion data and the astronomical parameters; The output feature is the vertical plane hourly total radiation.

[0028] Preferably, the input parameter combination needs to be selected according to different orientations, such as due east, due south, due west and due north, in order to fully consider the directional differences of solar radiation. This selection method ensures the applicability of the model in different orientations.

[0029] S4, using the selected input parameter combination, training a pre-defined machine learning model to obtain a trained vertical plane hourly total radiation calculation model.

[0030] In one implementation, an XGBoost model is used as the base model, and the base parameters of the model are set, including the maximum tree depth, the learning rate, the sub-sample ratio, the feature sampling ratio and the number of trees.

[0031] S5. Input the current multi-source data into the trained vertical surface hourly total radiation calculation model to obtain the vertical surface hourly total radiation calculation value of the target area.

[0032] It should be understood that the calculation results can be used for analysis or application, such as optimizing the arrangement of photovoltaic and solar thermal modules in building energy system design.

[0033] In one possible approach, after obtaining the predicted values, the hourly total radiance calculations for the vertical plane are visualized, including scatter plot visualizations and time series plot visualizations. For example, visualization tools such as Matplotlib or Tableau can be used to generate charts to help users intuitively understand the spatiotemporal distribution characteristics of the radiance data.

[0034] Through the above steps, this invention achieves a method for reliably predicting hourly total radiation of a vertical plane under different data conditions, improving the model's generalization ability in regions with scarce data. Those skilled in the art can make minor adjustments to the data source or model parameters according to actual needs, but all such adjustments fall within the protection scope of this invention.

[0035] Example 1 Combination Figure 2 This embodiment uses Xi'an as an example to specifically illustrate the calculation method of hourly total radiation in the vertical plane described in this invention. Those skilled in the art will understand that this embodiment is only for illustrating the invention and should not be considered as a limitation on the scope of protection of this invention.

[0036] Step S1: Obtain multi-source data for the target area.

[0037] Historical data for the Xi'an area was obtained, spanning four years (2016-2019), with a time resolution of one hour. The multi-source data includes: Horizontal solar radiation data are acquired through ground-based radiation observation instruments, including hourly total radiation on the horizontal surface (…). I h ), horizontal surface scattered radiation ( I d ); Surface meteorological data is obtained through surface meteorological stations, including hourly relative humidity ( ). RH h Hourly sunshine duration ( S h Hourly temperature T h ); Satellite inversion data can be obtained from third-party databases ( https: / / ceres.larc.nasa.gov / CERES-SYN1deg is used to obtain hourly cloud cover (CLOSE). N e ), hourly ozone (O 3) Hourly aerosol optical thickness ( AOD ) and hourly precipitation ( PW ); Astronomical parameters, including solar altitude angle, solar azimuth angle, solar incidence angle in each direction, declination angle, hour angle, and hourly astronomical radiation, are calculated using theoretical formulas.

[0038] Step S2: Determine the data status of the target area.

[0039] Based on the multi-source data obtained in step S1, the data status of the Xi'an area is determined as follows: If horizontal solar radiation data exists, it is determined that there is horizontal solar radiation data. If no horizontal solar radiation data is available, then the situation is determined to be horizontal solar radiation data.

[0040] In this embodiment, there is horizontal solar radiation data in the Xi'an area, so it is determined that there is horizontal solar radiation data.

[0041] Step S3: Preprocess the multi-source data.

[0042] In data preprocessing, time synchronization of multi-source data is performed to ensure consistency in the time series of hourly total radiation data and horizontal solar radiation data for each orientation. Furthermore, data quality control is implemented to remove outliers, such as data where the hourly total horizontal radiation exceeds the hourly astronomical solar radiation data, or data with a solar altitude angle less than 7 degrees.

[0043] Step S4: Select the combination of input parameters.

[0044] Step S41: Select the hourly total radiation in the directions of due east, due south, due west, and due north in sequence for the output features.

[0045] Step S42: Based on the data determined in step S2, select the input parameter combination. The input parameter combination is selected separately for different building orientations (east, south, west, north): For situations where radiation data is available, the selected input parameter combination should primarily consist of horizontal solar radiation data, for example: east-facing parameters should include hourly sunshine duration. S h ), total hourly radiation on the horizontal surface ( I h ); due west is the hourly total radiation of the horizontal plane ( I h Hourly sunshine duration ( S h ); The total hourly radiation on the horizontal plane is oriented due south. I h Hourly sunshine duration ( Sh ); the north orientation is the horizontal plane hourly total radiation (Gh) I h ), the normal hourly direct (Dh I b ), the hourly sunshine duration (Nsh S h ).

[0046] Step S5, constructing an XGBoost base model, setting the maximum tree depth, learning rate, sub-sample ratio, feature sampling ratio and the number of trees according to different orientations.

[0047] Specifically, the maximum tree depth and learning rate of XGBoost are 7 and 0.05; the sample sampling ratio, feature sampling ratio and the number of trees of the north orientation model are 0.8, 0.8 and 300 respectively; the sample sampling ratio, feature sampling ratio and the number of trees of the east orientation model are 0.8, 1.0 and 200 respectively; the sample sampling ratio, feature sampling ratio and the number of trees of the west orientation model are 1.0, 0.8 and 300 respectively; the sample sampling ratio, feature sampling ratio and the number of trees of the south orientation model are 0.8, 1.0 and 300 respectively.

[0048] Step S6, inputting the input parameters into the trained XGBoost model for testing to obtain the vertical plane hourly total radiation calculation value of the target area. Table 1 is the calculation error of each orientation hourly total radiation and the XGBoost model parameters under the condition of having radiation data in Xi'an area. From Table 1, it can be seen that the best input parameter combination of each orientation controls the number of input parameters while ensuring the accuracy, saving the calculation cost of the model.

[0049] Table 1 Vertical plane hourly total radiation model of different orientations under the condition of having radiation data in Xi'an area

[0050] The vertical plane hourly total radiation can be visualized. The prediction results are displayed through visualization, including: scatter plot visualization, drawing a scatter plot of the predicted value and the measured value, and calculating the determination coefficient (R 2 ), root mean square error (RMSE) and other indicators.

[0051] Time series visualization, taking time as the horizontal axis, drawing the fitting curve of the predicted value and the measured value, and displaying the prediction effect under different data conditions. The visualization results can be used to assist the scheduling decision and operation optimization of the solar power station.

[0052] Example two This example still takes Xi'an area as an example, assuming that the area has no horizontal plane solar radiation data condition.

[0053] Steps S1 to S3 are the same as in example one.

[0054] Step S4, select the output feature and input parameter combination.

[0055] Step S41, the output feature is selected as the hourly total radiation in the directions of due east, due south, due west and due north.

[0056] Step S42, according to the data condition determined in step S2, select the input parameter combination. The input parameter combination is selected for different building orientations (due east, due south, due west and due north) respectively: For the condition of no radiation and no meteorological data, the selected input parameter combination is mainly based on ground meteorological data, for example: for the due east and due west orientations, the selected input parameters are the solar elevation angle ( α ), hourly cloud cover ( N e ), hourly aerosol optical depth ( AOD ), hourly ozone ( O 3); for the due south orientation, the selected input parameters are the southward solar incident angle ( i s ), hourly cloud cover ( N e ); for the due north orientation, the selected input parameters are the horizontal plane hourly astronomical radiation ( I 0), hourly cloud cover ( N e ).

[0057] Step S5, construct the XGBoost base model, and set the maximum tree depth, learning rate, sub-sample ratio, feature sampling ratio and number of trees according to different orientations.

[0058] Specifically, the maximum tree depth and learning rate of XGBoost are 7 and 0.1 respectively; for the due north orientation model, the maximum tree depth, sample sampling ratio, feature sampling ratio and number of trees are 5, 0.8, 1.0 and 100 respectively; for the due south orientation model, they are 3, 0.8, 1.0 and 200 respectively; for the due west orientation model, they are 7, 1.0, 1.0 and 300 respectively; for the due east orientation model, they are 5, 1.0, 0.8 and 200 respectively.

[0059] Step S6, input the input parameters into the trained XGBoost model for testing to obtain the calculated value of the vertical plane hourly total radiation of the target area. Table 2 shows the calculation error of the hourly total radiation of each orientation under the condition of no radiation and no meteorological data in Xi'an area and the XGBoost model parameters.

[0060] Table 2 Hourly total radiation model of different orientations under the condition of no radiation and no meteorological data in Xi'an area

[0061] In order to better understand the prediction method of the hourly total radiation of each orientation, Figure 3The fitting graph of the prediction result and the measured value under the best input meteorological factor combination of each orientation in the area with radiation data. The results show that the priority introduction of solar radiation measurement data is an effective method. This correlation ranking-based strategy can not only improve the model accuracy in a very short time, but also reduce the number of input variables. Both of these aspects are of key significance in practical engineering applications. Figure 3 R 2 The determination coefficient (i.e., the correlation between the prediction and the actual) is represented as R, and the root mean square error is represented as RMSE.

[0062] As can be seen from the above embodiments, the vertical plane hourly total radiation data calculation method provided by the present application can effectively reduce the number of input parameters and the calculation cost on the one hand, and on the other hand, based on historical ground measured meteorological data, historical satellite inversion meteorological data and the prediction result obtained by the trained hourly total radiation prediction model of each orientation, the prediction accuracy can be improved, especially in the case of cloud cover and weather mutation, more accurate prediction results can be provided, so that the corresponding solar energy resources can be accurately predicted, and the solar power station can be better helped to be reasonably scheduled, and the power generation efficiency can be improved.

[0063] In another embodiment of the present application, a computer device is provided, which includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, and are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to realize the corresponding method flow or corresponding function; the processor in the embodiments of the present application can be used for the operation of the vertical plane hourly total radiation calculation method of the fusion of multi-source data.

[0064] In still another embodiment of the present application, the present application also provides a storage medium, specifically a computer readable storage medium (Memory), which is a memory device in a computer device, used for storing programs and data. It can be understood that the computer readable storage medium herein can include the built-in storage medium in the computer device, and of course can also include the expansion storage medium supported by the computer device. The computer readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory. One or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to realize the corresponding steps of the above-mentioned embodiment about the vertical plane hourly total radiation calculation method of fused multi-source data.

[0065] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) containing computer-usable program code.

[0066] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the flowcharts and / or block diagrams. Figure 1 The function specified in one or more flows and / or blocks Figure 1 The means for performing the function specified in one or more flows and / or blocks.

[0067] These computer program instructions can also be stored in a computer readable storage medium capable of directing the computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a product including instruction means, which implements the flowcharts and / or block diagrams. Figure 1 The function specified in one or more flows and / or blocksFigure 1 the function specified in the one or more blocks.

[0068] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable devices to generate a computer-implemented process, so that the instructions executed on the computer or other programmable devices provide a process for implementing the flowchart Figure 1 the flowchart or flowcharts and / or a block Figure 1 the steps of the function specified in the one or more blocks.

[0069] The present application also provides a computer program product, since the computer program product is used to execute any one of the above-mentioned fusion multi-source data vertical plane hourly total radiation calculation method. Since the computer program product provided by the present application belongs to the same inventive concept as the above-mentioned fusion multi-source data vertical plane hourly total radiation calculation method, the computer program product provided by the present application has all the advantages of the above-mentioned fusion multi-source data vertical plane hourly total radiation calculation method, so the beneficial effects of the computer program product provided by the present application will not be described one by one.

[0070] In the present application, the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0071] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit them, the protection scope of the present application is not limited thereto, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: any person skilled in the art within the technical range disclosed by the present application, it can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, all should be covered in the protection scope of the present application.

Claims

1. A method for calculating the hourly total radiation of a vertical plane by fusing multi-source data, characterized in that, include: Acquire multi-source data for the target area, including horizontal solar radiation data, ground meteorological data, satellite inversion data, and astronomical parameters; Based on the multi-source data, the data status of the target area is determined, including whether there is horizontal solar radiation data or no horizontal solar radiation data. Based on the data conditions, a corresponding combination of input parameters is selected from the multi-source data, wherein: when the data conditions are those with horizontal solar radiation data, the selected combination of input parameters includes the horizontal solar radiation data and ground meteorological data; when the data conditions are those without horizontal solar radiation data, the selected combination of input parameters includes the satellite inversion data and astronomical parameters; Using the selected combination of input parameters, a predefined machine learning model is trained to obtain a trained hourly total radiation calculation model for the vertical surface. The current multi-source data is input into the trained vertical surface hourly total radiation calculation model to obtain the calculated value of the vertical surface hourly total radiation of the target area.

2. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 1, characterized in that, The horizontal solar radiation data includes hourly total radiation and hourly scattered radiation on the horizontal surface. The ground meteorological data includes hourly relative humidity, hourly sunshine duration, and hourly temperature. The satellite inversion data includes hourly cloud cover, hourly ozone, hourly aerosol optical thickness, and hourly precipitation.

3. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 1, characterized in that, The determination of the data status of the target area specifically includes: If the aforementioned horizontal solar radiation data exists, it is determined that there is a horizontal solar radiation data situation. If the horizontal solar radiation data is not available, the condition is determined to be "no horizontal solar radiation data".

4. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 1, characterized in that, The input parameter combinations are selected separately for different building orientations, including due east, due south, due west, and due north.

5. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 1, characterized in that, The predefined machine learning model is the XGBoost model.

6. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 5, characterized in that, When training the XGBoost model, grid search or genetic algorithms are used to optimize the hyperparameters, which include maximum tree depth, learning rate, subsample ratio, feature sampling ratio, and number of trees.

7. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 1, characterized in that, After obtaining the hourly total radiation calculation value of the vertical plane of the target area, the method further includes: The hourly total radiation calculation value of the vertical plane is visualized, including scatter plot visualization and time series plot visualization.

8. The method for calculating the hourly total radiation of a vertical plane by fusing multi-source data according to claim 1, characterized in that, The astronomical parameters include solar altitude angle, solar azimuth angle, solar incidence angle in each direction, declination angle, hour angle, and astronomical radiation.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a method for calculating the hourly total radiation of a vertical plane by fusing multi-source data as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a method for calculating the hourly total radiation of a vertical plane by fusing multi-source data as described in any one of claims 1 to 8.

Images