A high spatio-temporal resolution photovoltaic resource prediction system along highways

Abstract

The present invention discloses a photovoltaic resource prediction system with high spatio-temporal resolution along highways. The system is based on the WRF-Chem-Solar model, combines CMIP6 meteorological data and DPEC emission source data to simulate future weather and climate, and uses radiation spatial downscaling and terrain correction technologies to establish a refined assessment and prediction method for highway photovoltaic resources, obtaining solar energy-related meteorological prediction data such as global radiation, horizontal and tilted direct radiation, diffuse radiation, and air temperature with high spatio-temporal resolution along highways. And through PVLIB-Python, the resource assessment results are modeled and analyzed to obtain photovoltaic development potential assessment and prediction products along highways, realizing refined assessment and accurate prediction of photovoltaic resources with high spatio-temporal resolution along national highways.

Description

Technical Field

[0001] The present invention belongs to the technical field of meteorological science and technology, and particularly relates to a photovoltaic resource prediction system with high spatio-temporal resolution along highways. Background Art

[0002] The development and utilization of distributed photovoltaic resources that can be used immediately upon generation provide a new power supply approach to meet the increasing electricity demand of highways. Moreover, with the rapid development of photovoltaic technology, the efficient development and utilization of solar energy resources along highways have broad development space and huge application potential.

[0003] The highway photovoltaic resource assessment and prediction system aims at the fine assessment of highway photovoltaic resources and the prediction of future solar power generation. Considering the influence of clouds and aerosols on solar radiation, based on the carbon neutrality target emission source inventory, historical and future global climate projection data, it adopts a meteorological-chemical coupling model and an improved cloud microphysical parameterization scheme to achieve the assessment and prediction of solar radiation and power generation with high spatio-temporal resolution along highways in the next few decades.

[0004] In addition, the system also combines radiation spatial downscaling technology, which can improve the resolution of solar radiation data to a level sufficient to cover a single highway section, which is crucial for the refined assessment of photovoltaic resources. Through this high-resolution assessment, the potential and economic benefits of building solar power generation facilities on specific sections can be determined more accurately. At the same time, the system also considers the influence of terrain. Through terrain correction technology, the photovoltaic resources under complex terrain conditions along highways can be evaluated more accurately. This is particularly important for mountainous or terrain-complex highway sections. Through this terrain correction, the accuracy of photovoltaic resource assessment can be ensured, thus providing more reliable data support for the design and layout of solar power generation facilities.

[0005] The system can provide important scientific and technological support for the fine assessment of highway photovoltaic resources and the prediction of future power generation. This is of great significance for promoting the green energy transformation of transportation infrastructure such as highways and accelerating the achievement of the zero-carbon target in the transportation industry. Summary of the Invention

[0006] Objective of the Invention: Aiming at the problems existing in the prior art, the present invention provides a photovoltaic resource prediction system with high spatio-temporal resolution along highways. Based on the WRF-Chem-Solar model, combined with CMIP6 meteorological data and DPEC emission source data, it simulates future weather and climate, and uses radiation spatial downscaling and terrain correction technologies to establish a fine evaluation and prediction method for highway photovoltaic resources, obtaining solar-related meteorological prediction data such as global radiation, horizontal and tilted direct radiation, diffuse radiation, and air temperature with high spatio-temporal resolution along highways. And through PVLIB-Python, it conducts modeling analysis on the resource evaluation results to obtain products for evaluating and predicting the photovoltaic development potential along highways, realizing fine evaluation and accurate prediction of photovoltaic resources with high spatio-temporal resolution along national highways.

[0007] Technical Solution: To achieve the above objective of the invention, the present invention adopts the following technical solutions: A photovoltaic resource prediction system with high spatio-temporal resolution along highways, including the following steps:

[0008] S1, Data Preparation: Obtain DPEC emission source data and CMIP6 meteorological field data;

[0009] S2, Data Processing:

[0010] (1) Processing of DPEC Emission Source Data: Conduct hourly three-dimensional grid processing on DPEC emission sources to form three-dimensional grid values reflecting the spatio-temporal characteristics of various emission sources;

[0011] (2) Processing of CMIP6 Meteorological Field Data: The CMIP6 meteorological field data uses 18 CMIP6 models, namely ACCESS-CM2, ACCESS-ESM1–5, CanESM5, BCC-CSM2-MR, FGOALS-f3-L, FGOALS-g3, EC-Earth3, EC-Earth3-Veg, IPSL-CM6A-LR, AWI-CM-1-1-MR, MPI-ESM1-2-HR, MPI-ESM1-2-LR, MIROC6, MRI-ESM2-0, NorESM2-LM, CESM2, CESM2-WACCM, GFDL-ESM4, and conducts integration and three-dimensional grid conversion processing on the 18 CMIP6 models;

[0012] S3, Constructing the WRF-Chem-Solar Model: Based on the WRF-Chem-Solar model, and conduct the following settings and operations on the model:

[0013] (1) Setting of Model Parameters: Adopt a double nested grid model. The first layer of nesting uses a resolution of 50 km or higher, and a resolution of 10 km or higher is used for studying local areas;

[0014] (2) Microphysical parameter settings: The Dudhia microphysical model is used to calculate the mutual conversion process of hydrometeors including cloud water, rain water, snow, and hail; the CBMZ model is used for cloud convection parameterization calculation; the MOSAIC model is used to calculate the physical properties and interactions of hydrometeors in different particle size segments;

[0015] (3) Initial field settings: The CMIP6 meteorological field data uses the current simulation data as the initial condition; the chemical field of the DPEC emission source uses the simulation data of the last time step of the previous day as the initial field condition;

[0016] (4) WRF-Chem-Solar model calculation scheme settings: The model uses 24 hours before the start of the simulation as the spin-up time and the subsequent 24 hours as the output of the simulation results to obtain a gridded meteorological product of future solar energy resources with high spatio-temporal resolution;

[0017] S4, Post-processing of the WRF-Chem-Solar model: For the gridded meteorological product obtained in step S3, interpolation processing is performed through the SES2 radiative transfer model to achieve radiative spatial downscaling;

[0018] S5, Terrain correction: Based on the solar radiation correction model under complex terrain of DEM along the highway, the spatio-temporal distribution of solar radiation under complex terrain conditions along the highway is calculated;

[0019] S6, Photovoltaic potential assessment and prediction: Using the PVLIB-Python numerical simulation system, the meteorological data after radiative spatial downscaling and terrain correction is used for photovoltaic system performance modeling and analysis to obtain photovoltaic potential assessment and prediction products along the highway under different tilts, different tracking methods, and different photovoltaic materials.

[0020] Furthermore, in step S1, the DPEC emission source data uses the Ambitious-pollution-Neutral-goal scenario, which is one of the six emission scenarios in the second set of scenario datasets of the DPEC model; CMIP6 uses the SP1-2.6 scenario.

[0021] Furthermore, in step S2, after the DPEC emission source data is three-dimensionally gridded, the data resolution is 10 km * 10 km, and there are 41 layers in the vertical direction, with a layer height of 2 km.

[0022] Furthermore, the interpolation process of the SES2 radiative transfer model in step S4 includes:

[0023] (1) Horizontal direction downscaling interpolation:

[0024] Search for and determine the point to be interpolated The four nearest neighbor points around, and the latitude and longitude coordinates of the four nearest neighbor points are , , , , and the value of the point to be interpolated with horizontal downscaling at is calculated by the following formula ,

[0025] ,

[0026] In the formula, , , and respectively represent the values of the four nearest neighbor points, represents the value of the coordinate point , represents the value of the coordinate point ;

[0027] (2) Vertical downscaling interpolation:

[0028] For the input variables of the WRF-Chem-Solar model, linear interpolation method is adopted in the vertical direction, and the variable value T of each input variable at the pressure layer to be interpolated is calculated by the following formula

[0029] ,

[0030] In the formula, T represents the variable value of the pressure layer to be interpolated, p represents the pressure value of the pressure layer to be interpolated, T a represents the variable value of the ath pressure layer, T b represents the variable value of the bth pressure layer, p a represents the pressure value of the ath pressure layer, p b represents the pressure value of the bth pressure layer; the input variables include temperature, pressure, ozone concentration, carbon dioxide concentration, nitric oxide concentration, methane concentration and oxygen concentration;

[0031] For pressures above 100 mb, the water vapor mixing ratio is replaced by the mid-latitude atmospheric climate profile; for pressures below 100 mb, the water vapor mixing ratio qq _ need m of the interpolation layer is calculated by the following formula

[0032] ,

[0033] In the formula, the subscript i represents the layer number before interpolation, the subscript m represents the layer number of the interpolation layer, P i represents the pressure of the ith layer, P i-1 represents the pressure of the (i - 1)th layer, P i+1Denotes the air pressure at the (i + 1)-th layer, Denotes the difference in air pressure between the layer before interpolation and the previous layer, vapor i-1 Denotes the water vapor content at the (i - 1)-th layer, qq i Denotes the cumulative water vapor amount from the ground to the i-th layer, qq i-1 Denotes the cumulative water vapor amount from the ground to the (i - 1)-th layer, qq1 i-1 Denotes the cumulative water vapor amount at the (i - 1)-th layer; qq m Denotes the cumulative water vapor amount from the ground to the interpolation layer, qq i+1 Denotes the cumulative water vapor amount at the (i + 1)-th layer, P m Denotes the air pressure at the interpolation layer, P m+1 Denotes the air pressure at the layer above the interpolation layer, qq _ need m Denotes the water vapor mixing ratio at the interpolation layer, qq m+1 Denotes the cumulative water vapor amount from the ground to the layer above the interpolation layer.

[0034] Furthermore, in step S2, the 18 CMIP6 models are integrated and calculated through the following formula to generate the integrated and corrected GCM data GCM mvt ,

[0035] ,

[0036] In the formula, GCM mvt Denotes the global climate model after integrated correction; GCM denotes the global climate model; GCM’ denotes the interannual perturbation term of the global climate model GCM;

[0037] Denotes the average value of the long-term trend LT of the reanalysis data in the historical period H;

[0038] Denotes the average value of the long-term trend LT of the multi-model ensemble MME in the historical period H;

[0039] Denotes the average deviation of the GCM data from the long-term trend of the reanalysis dataset in the historical period;

[0040] Denotes the standard deviation of the detrended reanalysis data within the historical period, Denotes the standard deviation of the detrended GCM data; r s Denotes the ratio of the standard deviation of the detrended reanalysis data to the standard deviation of the detrended GCM data within the historical period.

[0041] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0042] (1) The system can predict the photovoltaic resources at specific future times, and obtain photovoltaic resource prediction products with high spatio-temporal resolution (hourly or every 15 minutes, at a scale of 100 meters or dozens of meters) along the highway; and consider the impacts of different tilts, different tracking methods, and different photovoltaic materials on the prediction of photovoltaic potential.

[0043] (2) The system includes a cloud-aerosol-radiation interaction feedback mechanism, and fully considers the impacts of clouds and aerosols on solar radiation.

[0044] (3) The system adopts a radiation spatial downscaling technique. Compared with the traditional WRF model, it has a higher resolution and can easily obtain solar radiation data at a scale of 100 meters or dozens of meters along the highway.

[0045] (4) The system adopts a terrain correction technique. Compared with the traditional planar model, it fully considers the impacts of the complex terrain along the highway and can more accurately evaluate and predict the photovoltaic potential under the complex terrain conditions along the highway. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of the photovoltaic resource prediction system with high spatio-temporal resolution along the highway described in the present invention;

[0047] Figure 2 It is a schematic diagram of bilinear interpolation according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0048] The following further clarifies the present invention in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. After reading the present invention, various equivalent modifications made by those skilled in the art to the present invention all fall within the scope defined by the appended claims of this application.

[0049] As Figure 1 shown, the present invention adopts the following solution to obtain fine evaluation and prediction products of photovoltaic resources with high spatio-temporal resolution in the next few decades or hundreds of years along the highway, and realizes accurate estimation of photovoltaic resources.

[0050] S1, Data preparation:

[0051] ① DPEC emission source: The present invention adopts the Ambitious-pollution-Neutral-goal scenario, which is one of the six emission scenarios in the second set of scenario datasets (v1.1) of the DPEC model, that is, the carbon neutral climate scenario.

[0052] ②CMIP6 meteorological field data: CMIP6 (Coupled Model Intercomparison Project Phase6) is the sixth phase of the global climate model intercomparison project. CMIP6 data covers multiple levels from the ground to the atmosphere, including multiple meteorological variables such as temperature, precipitation, wind speed, humidity, etc. The SSP1-2.6 scenario is adopted in this system.

[0053] S2. Treatment of DPEC emission sources for the WRF-Chem-Solar model:

[0054] The WRF-Chem-Solar model is a numerical weather prediction model specifically designed for solar energy resource assessment and forecasting needs. It is based on the WRF-Solar model and further combined with the WRF-Chem model. In an online coupling manner, the aerosol chemistry module is combined with the radiation module.

[0055] Based on the DPEC emission sources, the emission source treatment mode of the WRF-Chem-Solar model is used for localization processing to support a more refined three-dimensional grid tracing mode with higher resolution. The purpose of the processing is to provide hourly three-dimensional grid emission sources. The horizontal resolution of the grid can reach 10km * 10km or higher, and there are 41 layers in the vertical direction, mainly concentrated below 2km.

[0056] S3. Treatment of CMIP6 large-scale meteorological fields:

[0057] The meteorological field data of this invention uses the integrated prediction data of 18 CMIP6 models, namely ACCESS-CM2, ACCESS-ESM1–5, CanESM5, BCC-CSM2-MR, FGOALS-f3-L, FGOALS-g3, EC-Earth3, EC-Earth3-Veg, IPSL-CM6A-LR, AWI-CM-1-1-MR, MPI-ESM1-2-HR, MPI-ESM1-2-LR, MIROC6, MRI-ESM2-0, NorESM2-LM, CESM2, CESM2-WACCM, GFDL-ESM4, a total of 18 models. It transcends the limitations of a single model, and the meteorological prediction effect is better than that of a single model, with stronger reliability. The WPS (WRF Preprocessing System) module of the WRF-Chem-Solar model converts the CMIP6 meteorological field data into the three-dimensional grid data required by the model, and these data are subsequently used as the initial field and boundary conditions of the WRF-Chem-Solar model.

[0058] The 18 CMIP6 models of the present invention are based on the WPS module, and the integrated corrected GCM data GCM is generated through the following formula for integrated calculation mvt ,

[0059] ,

[0060] In the formula, GCM mvt represents the global climate model after integrated correction; GCM represents the global climate model; GCM’ represents the interannual perturbation term of the global climate model GCM; represents the average value of the long-term trend LT of the reanalysis data in the historical period H; represents the average value of the long-term trend LT of the multi-model ensemble MME in the historical period H; represents the average deviation of the GCM data from the long-term trend of the reanalysis dataset in the historical period; represents the standard deviation of the detrended reanalysis data within the historical period, represents the standard deviation of the detrended GCM data; r s represents the ratio of the standard deviation of the detrended reanalysis data to the standard deviation of the detrended GCM data within the historical period.

[0061] The present invention uses the integrated prediction data of 18 CMIP6 models to replace the conventional single CMIP6 model, and the prediction effect is better than that of the single model.

[0062] S4. Use the WRF-Chem-Solar model to predict the meteorological field in the next few decades or hundreds of years:

[0063] (1) Model parameter setting: Based on the WRF-Chem-Solar model, the model adopts the dual nested grid technology. The first layer of nesting is at a resolution of 50 km or higher, and the second layer of nesting is further improved to a resolution of 10 km or higher, with significantly improved spatio-temporal resolution;

[0064] (2)Mode parameter settings: In terms of the microphysical parameterization scheme, the mode of the present invention has also been significantly improved. The parameterization scheme selected is Scheme D + CBMZ + MOSAIC 4-bin. Scheme D is the Dudhia microphysical scheme, which mainly considers the mutual conversion processes of hydrometeors such as cloud water, rain, snow, and hail. The CBMZ scheme refers to an improved version of the Kain-Fritsch cumulus convection parameterization scheme (KF), namely the revised version (CBMZ) of the Kain-Fritsch cumulus convection parameterization scheme. It is mainly used to describe the cumulus convection process, and by parameterizing the physical mechanism of cumulus convection, it simulates the development of convective clouds and precipitation formation. The MOSAIC scheme is a bin microphysical parameterization scheme, and 4-bin means that the hydrometeors are divided into 4 particle size segments for simulation. It more precisely simulates the microphysical processes of clouds and precipitation by detailing the physical characteristics and their interactions of hydrometeors in different particle size segments. Among them, the Morrison microphysical scheme can replace Kessler, and on the basis of the WRF-Chem 4.2 version, the Morrison scheme has been further optimized. The cloud droplet activation process is implanted in the Morrison scheme and the cloud droplet number concentration is updated in real time, improving cloud physical parameters such as cloud droplet mass concentration, cloud droplet radius, and cloud water path, and thus improving the simulation of the cloud-radiation feedback process;

[0065] (3)Initial field settings: The CMIP6 meteorological field data uses the current simulation data as the initial condition; the chemical field of the DPEC emission source uses the simulation data of the last time step of the previous day as the initial field condition;

[0066] (4)In the calculation scheme, the WRF-Chem-Solar mode conducts simulations one day in advance every day of calculation. The first 24h of the simulation is used as the spin-up time, and the subsequent 24 hours are used as the simulation results.

[0067] In the present invention, in the settings of the initial field and boundary conditions, the meteorological field adopts CMIP6 cold start, and the chemical field uses the simulation results of the last time step of the previous day as the initial field. Such settings enable the mode to more accurately combine the meteorological field and the chemical field, and further consider the interactions among cloud-aerosol-radiation. Finally, gridded meteorological products related to solar energy resources with high spatio-temporal resolution (hourly or every 15 minutes, 3 kilometers or finer) from 2015 to 2060 or longer are simulated.

[0068] Through the above parameter settings, the present invention can predict the spatio-temporal changes of solar energy resources in a certain month, season, year, multiple years, or longer in the future; the microphysical parameterization scheme has been locally improved, and the interactions among cloud-aerosol-radiation have been fully considered, significantly improving the spatio-temporal resolution.

[0069] S5, Radiation spatial downscaling based on the SES2 radiative transfer model:

[0070] In the present invention, interpolation processing is carried out through the SES2 radiative transfer model to achieve radiation spatial downscaling. The data output by the WRF-Chem-Solar model cannot be directly applied to the SES2 model. The SES2 model requires 91 layers, while the WRF-Chem-Solar model only outputs 34 layers of data. Interpolation of the model output data is required before it can be used. The present invention performs interpolation processing according to the following method:

[0071] (1) Horizontal direction downscaling interpolation

[0072] In the process of horizontal direction downscaling, this solution adopts the method of bilinear interpolation. By using the values of four stations near the grid points, linear interpolation is performed in the horizontal and vertical directions to obtain the value of the grid point. This method shows good results in spatial downscaling, especially in the refined forecasting of precipitation data. The process of bilinear interpolation is as follows:

[0073] ① Determine the nearest neighbor points: First, determine the four nearest neighbor points around the interpolation point , , , , which form a small rectangular area, and the interpolation point P is located inside this rectangle;

[0074] ② Linear interpolation: Perform linear interpolation in the x direction and y direction respectively. For the x direction, use the two endpoint values R1 and R2 of the row where the interpolation point is located for linear interpolation; for the y direction, use the two endpoint values of the column where the interpolation point is located for linear interpolation, as Figure 2 shown. Calculate the interpolation point P, that is, the value at the point to be interpolated through the following formula ,

[0075] ,

[0076] In the formula, , , and respectively represent the values of the four nearest neighbor points, represents the value of the coordinate point , represents the value of the coordinate point ;

[0077] (2) Vertical direction downscaling interpolation

[0078] In the vertical downscaling process, input variables such as carbon dioxide, nitrous oxide, methane, oxygen, ozone, temperature, specific humidity, cloud liquid water content, and cloud ice crystal content are used. Among them, carbon dioxide, nitrous oxide, methane, oxygen, and ozone are set as constant values. In addition, since there will be a large error in using the extrapolation method for water vapor at 100 mb, in the present invention, for water vapor and temperature above 100 mb, the mid-latitude atmospheric climate profile is used instead, and for below 100 mb, the output data of the WRF-Chem model is adopted, specifically as follows:

[0079] The variable value T of each input variable at the pressure layer to be interpolated,

[0080] ,

[0081] In the formula, T represents the variable value of the pressure layer to be interpolated, p represents the pressure value of the pressure layer to be interpolated, T a represents the variable value of the a-th pressure layer, T b represents the variable value of the b-th pressure layer, p a represents the pressure value of the a-th pressure layer, p b represents the pressure value of the b-th pressure layer; the input variables include temperature, pressure, ozone concentration, carbon dioxide concentration, nitric oxide concentration, methane concentration, and oxygen concentration;

[0082] For pressures above 100 mb, the water vapor mixing ratio is replaced by the mid-latitude atmospheric climate profile; for pressures below 100 mb, the water vapor mixing ratio qq _ need m is calculated by the following formula,

[0083] ,

[0084] In the formula, the subscript i represents the layer number before interpolation, the subscript m represents the layer number of the interpolation layer, P i represents the pressure of the i-th layer, P i-1 represents the pressure of the (i - 1)-th layer, P i+1 represents the pressure of the (i + 1)-th layer, represents the pressure difference between the layer number before interpolation and the upper layer, vapor i-1 represents the water vapor content of the (i - 1)-th layer, qq i represents the cumulative water vapor amount from the ground to the i-th layer, qq i-1 represents the cumulative water vapor amount from the ground to the (i - 1)-th layer, qq1 i-1 represents the cumulative water vapor amount of the (i - 1)-th layer; qq m represents the cumulative water vapor amount from the ground to the interpolation layer, qq i+1 represents the cumulative water vapor amount of the (i + 1)-th layer, P m represents the pressure of the interpolation layer, P m+1Indicates the air pressure of the previous layer of the interpolation layer, qq _ need m represents the water vapor mixing ratio of the interpolation layer, qq m+1 Represents the accumulated water vapor from the ground to the previous layer of the interpolation layer.

[0085] After interpolation of the model output data, the temperature, air pressure, water vapor, ozone, aerosol, cloud water, cloud ice, cloud vertical distribution profile, surface reflectivity, and the real-time calculation data of the solar zenith angle of the SES2 model and the background data of greenhouse gases such as carbon dioxide, nitric oxide, methane, oxygen, CFC, etc. are input into the SES2 model. The SES2 model undergoes preprocessing, vertical interpolation, diagnostic quantity calculation, solar radiation calculation, photovoltaic resource prediction, and ground temperature and reflectivity calculation. Photovoltaic resource prediction products with a resolution of hundreds or tens of meters along the highway are obtained, including horizontal or inclined surface total radiation, direct radiation, scattered radiation, ground temperature, etc. in the narrow band (0.25~1.1μm) of photovoltaic power generation.

[0086] Compared with the traditional WRF model, the present invention further combines the radiation spatial downscaling technology to significantly improve the spatiotemporal resolution, and can obtain photovoltaic resource data at a scale of hundreds or tens of meters along the highway at a specific time in the future.

[0087] S6, terrain correction of photovoltaic resource prediction products along highways

[0088] Compared with horizontal plane radiation, solar radiation along the highway is significantly affected by the terrain, which is mainly due to the differences in the reception of direct sunlight, atmospheric scattering and surrounding surface reflected radiation caused by the undulating terrain. In order to perform effective terrain correction, the present invention adopts the solar radiation correction model under complex terrain using DEM.

[0089] The present invention takes into account the self-shading caused by slope, aspect and terrain, and also considers the anisotropy of radiation and the influence of slope reflected radiation on the total solar radiation under complex terrain in the model, and calculates the temporal and spatial distribution of solar radiation under complex terrain conditions along the highway. By calculating the following formula,

[0090] ,

[0091] In the formula, represents direct solar radiation, represents the solar diffuse radiation, Indicates that the terrain reflects radiation, while the sun directly radiates By calculating the following formula,

[0092] ,

[0093] In the formula, is the total direct radiation on the horizontal plane, is the extraterrestrial radiation on the horizontal plane, is the extraterrestrial radiation under complex terrain conditions, and the unit is MJ / m 2 ; T represents the duration of each day, is the sun-earth correction coefficient, k represents the number of sunshine hours, represents the solar constant, and respectively represent the initial hour angle and the final hour angle of sunshine hours under complex terrain conditions, represents the solar declination, u represents the north-south slope coefficient of the terrain, v represents the east-west slope coefficient of the terrain, and w represents the vertical slope coefficient of the terrain.

[0094] The solar diffuse radiation on complex terrain is calculated by the following formula,

[0095] ,

[0096] In the formula, K t represents the clear sky index, K b represents the direct transmittance, and R b represents the ratio of extraterrestrial radiation under complex terrain conditions to extraterrestrial radiation on the horizontal plane. V represents the terrain openness, and the terrain openness at any point is calculated by the following formula,

[0097] ,

[0098] In the formula, a i is the maximum elevation angle of the point in one direction, that is, the maximum shelter degree of the point in that direction; n is the number of azimuth angles. In the present invention, the azimuth angle step is taken as 5°, and there are 72 in a full circle. The radius of the shelter range is 20 km.

[0099] The solar reflected radiation projected from the surrounding terrain received by the mountain depends on the mountain albedo and the terrain openness, and is calculated by the following formula,

[0100] ,

[0101] In the formula, Q represents the total solar radiation on the horizontal plane, and the unit is MJ / m 2 , a s is the surface albedo.

[0102] Compared with the traditional plane model, the present invention fully considers the influence of complex terrain along the highway, can more accurately evaluate and predict the photovoltaic potential under complex terrain conditions along the highway, and the product is more reliable.

[0103] S7, Precise Evaluation and Prediction of Photovoltaic Potential along Expressways: Using the PVLIB-Python numerical simulation system, the meteorological data with radiation spatial downscaling and terrain correction (i.e., the photovoltaic resource prediction products with a resolution of one hundred meters or dozens of meters along the expressway obtained in the above steps) are input into the photovoltaic system performance modeling and analysis, so as to obtain the evaluation and prediction products of photovoltaic potential along the expressway under different tilting angles, different tracking modes, and different photovoltaic materials. The above photovoltaic system performance modeling and analysis include solar irradiance calculation, photovoltaic module and system performance simulation, photovoltaic system modeling, shadow analysis, etc.

[0104] Inventive Point: Considering the influence of different tilting angles, different tracking modes, and different photovoltaic materials on the evaluation and prediction of photovoltaic potential.

[0105] The above are only partial embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A photovoltaic resource prediction system with high spatio-temporal resolution along highways, characterized in that It includes the following steps: S1. Data preparation: Obtain DPEC emission source data and CMIP6 meteorological field data; S2. Data processing: (1) DPEC emission source data processing: Conduct hourly three-dimensional grid processing on DPEC emission sources to form three-dimensional grid values reflecting the spatio-temporal characteristics of various emission sources; (2) CMIP6 meteorological field data processing: The CMIP6 meteorological field data uses 18 CMIP6 models, namely ACCESS-CM2, ACCESS-ESM1–5, CanESM5, BCC-CSM2-MR, FGOALS-f3-L, FGOALS-g3, EC-Earth3, EC-Earth3-Veg, IPSL-CM6A-LR, AWI-CM-1-1-MR, MPI-ESM1-2-HR, MPI-ESM1-2-LR, MIROC6, MRI-ESM2-0, NorESM2-LM, CESM2, CESM2-WACCM, GFDL-ESM4, and perform integration and three-dimensional grid conversion processing on the 18 CMIP6 models; S3. Construct the WRF-Chem-Solar model: Based on the WRF-Chem-Solar model, perform the following settings and operations on the model: (1) Model parameter settings: Adopt a double nested grid model. The first layer of nesting uses a resolution of 50 km or higher, and a resolution of 10 km or higher is used for studying local areas; (2) Microphysical parameter settings: Use the Dudhia microphysical model to calculate the mutual conversion process of hydrometeors including cloud water, rain water, snow, and hail; use the CBMZ model to calculate cloud convection parameterization; use the MOSAIC model to calculate the physical properties and interactions of hydrometeors in different particle size segments; (3) Initial field settings: Use the current simulation data of CMIP6 meteorological field data as the initial condition; use the simulation data of the last time step of the previous day as the initial field condition for the chemical field of DPEC emission sources; (4) WRF-Chem-Solar model calculation scheme settings: The model uses 24 h before the start of the simulation as the spin-up time and the subsequent 24 h as the output of the simulation results to obtain a grid-based meteorological product of future solar energy resources with high spatio-temporal resolution; S4. Post-processing of the WRF-Chem-Solar model: Interpolate the grid-based meteorological product obtained in step S3 through the SES2 radiative transfer model to achieve radiative spatial downscaling; S5. Terrain correction: Based on the solar radiation correction model under complex terrain of DEM along the highway, calculate the spatio-temporal distribution of solar radiation under complex terrain conditions along the highway; S6. Photovoltaic potential assessment and prediction: Use the PVLIB-Python numerical simulation system to apply the meteorological data after radiative spatial downscaling and terrain correction to the performance modeling and analysis of the photovoltaic system, and obtain the photovoltaic potential assessment and prediction products along the highway under different tilts, different tracking methods, and different photovoltaic materials.

2. The high-temporal and high-spatial resolution photovoltaic resource prediction system along the highway according to claim 1, wherein: In step S1, the DPEC emission source data adopts the Ambitious-pollution-Neutral-goal scenario, which is one of the six emission scenarios in the second set of scenario data of the DPEC model; CMIP6 adopts the SP1-2.6 scenario.

3. The high-temporal and high-spatial resolution photovoltaic resource prediction system along the highway according to claim 1, wherein: In step S2, after the three-dimensional gridification of the DPEC emission source data, the data resolution is 10 km * 10 km, and there are 41 layers in the vertical direction, with a layer height of 2 km.

4. The high spatio-temporal resolution photovoltaic resource prediction system along the highway according to claim 1, characterized in that: In step S4, the interpolation process of the SES2 radiative transfer model includes: (1) Downscaling interpolation in the horizontal direction: Search for and determine the four nearest neighbor points around the point to be interpolated whose longitude and latitude coordinates are , , , , and the value of the point to be interpolated for horizontal downscaling at is calculated by the following formula ， Wherein, , , and respectively represent the values of four nearest neighbor points, represents the value of the coordinate point ; represents the value of the coordinate point ; (2) Downscaling interpolation in the vertical direction: For the input variables of the WRF-Chem-Solar model, the linear interpolation method is adopted in the vertical direction, and the variable value T of each input variable at the pressure layer to be interpolated is calculated by the following formula. ， Wherein, T represents the variable value of the pressure layer to be interpolated, p represents the pressure value of the pressure layer to be interpolated, T a represents the variable value of the a-th pressure layer, T b represents the variable value of the b-th pressure layer, p a represents the pressure value of the a-th pressure layer, p b represents the pressure value of the b-th pressure layer; the input variables include temperature, pressure, ozone concentration, carbon dioxide concentration, nitric oxide concentration, methane concentration, and oxygen concentration; For pressures above 100 mb, the water vapor mixing ratio is replaced by the mid-latitude atmospheric climate profile; for pressures below 100 mb, the water vapor mixing ratio qq of the interpolation layer _ need m is calculated by the following formula ， In the formula, the subscript i represents the layer number before interpolation, the subscript m represents the layer number of the interpolation layer, and P i represents the air pressure of the i-th layer, and P i-1 represents the air pressure of the (i - 1)-th layer, and P i+1 represents the air pressure of the (i + 1)-th layer, represents the air pressure difference between the layer number before interpolation and the upper layer, vapor i-1 represents the water vapor content of the (i - 1)-th layer, qq i represents the cumulative water vapor amount from the ground to the i-th layer, qq i-1 represents the cumulative water vapor amount from the ground to the (i - 1)-th layer, qq1 i-1 represents the cumulative water vapor amount of the (i - 1)-th layer; qq m represents the cumulative water vapor amount from the ground to the interpolation layer, qq i+1 represents the cumulative water vapor amount of the (i + 1)-th layer, P m represents the air pressure of the interpolation layer, P m+1 represents the air pressure of the layer above the interpolation layer, qq _ need m represents the water vapor mixing ratio of the interpolation layer, qq m+1 represents the cumulative water vapor amount from the ground to the layer above the interpolation layer.

5. The photovoltaic resource prediction system with high spatio-temporal resolution along the highway according to claim 1, characterized in that: In step S2, the 18 CMIP6 models are integrated and calculated through the following formula to generate the GCM data GCM after integrated correction mvt , ， where GCM mvt represents the global climate model after integrated correction; GCM represents the global climate model; GCM’ represents the interannual perturbation term of the global climate model GCM; represents the average value of the long-term trend LT of the reanalysis data over the historical period H; represents the average value of the long-term trend LT of the multi-model ensemble MME over the historical period H; Indicates the average deviation of GCM data from the reanalysis dataset in the long-term trend during the historical period; represents the standard deviation of the detrended reanalysis data within the historical period, represents the standard deviation of the detrended GCM data; r s represents the ratio of the standard deviation of the detrended reanalysis data to the standard deviation of the detrended GCM data within the historical period.

Images