Complex terrain pollutant concentration prediction method based on WRF and CFD coupling

Abstract

The invention provides a complex terrain pollutant concentration prediction method based on WRF and CFD coupling, and the method specifically comprises the following steps: constructing a three-dimensional space grid of a complex terrain, generating meteorological element data covering a computational domain of a CFD mode, building a dynamic coupling mechanism of a WRF mode and the CFD mode, and carrying out the numerical calculation based on the CFD. And flow field data of the pollutant diffusion process under the complex terrain condition are obtained. According to the complex terrain pollutant concentration prediction method provided by the invention, the macroscopic meteorological background of WRF and the microscopic terrain effect of CFD are deeply fused, seamless connection simulation from large-scale circulation to small-scale turbulence is realized, and the temporal-spatial resolution and reliability of pollutant concentration prediction in a complex environment are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of atmospheric environment, in particular to the field of pollutant concentration prediction, and specifically to a complex terrain pollutant concentration prediction method based on WRF and CFD coupling. BACKGROUND

[0002] The diffusion process of pollutants in the atmosphere is significantly affected by meteorological conditions and topographic factors. In complex terrain areas, air flow movement exhibits highly nonlinear characteristics, such as mountain flow, climbing effect, etc. These factors lead to complex and variable pollutant diffusion paths and concentration distributions. Complex meteorological conditions such as wind speed and direction determine the speed and direction of pollutant diffusion, temperature and inversion layer affect vertical diffusion capacity, atmospheric turbulence promotes pollutant mixing and dilution, etc. Each factor is interrelated and affects the diffusion process and spatial and temporal distribution of pollutants in the atmosphere.

[0003] Currently, pollutant concentration prediction methods mainly include the Weather Research and Forecasting (WRF) model and the Computational Fluid Dynamics (CFD) model. The WRF model, as a typical mesoscale meteorological model, can simulate the evolution of large-scale meteorological fields and output meteorological elements such as wind speed, wind direction, temperature, and air pressure, providing a macro meteorological background for pollutant diffusion simulation. However, the spatial resolution of the WRF model is usually on the order of kilometers, making it difficult to accurately depict small-scale complex terrain or air flow disturbances around building groups, and unable to meet the needs of fine prediction of local area pollutant concentration. The CFD method solves the Navier-Stokes (N-S) equation and other fluid mechanics control equations to simulate small-scale complex flow fields with high accuracy, accurately capturing phenomena such as vortices, air flow separation and reattachment around buildings, and then simulating the diffusion process of pollutants in detail. However, the CFD method has high computational cost when simulating large areas, and the setting of boundary conditions relies on experience or simplification assumptions, making it difficult to accurately reflect large-scale meteorological conditions in the real atmospheric environment.

[0004] Some existing numerical simulation methods use only the WRF model or the CFD method, which cannot simultaneously consider the accurate simulation of large-scale meteorological background and small-scale complex terrain flow field. Some studies attempt to combine WRF and CFD for pollutant concentration prediction, but most use a simple coupling method, only using the fitting function curve of meteorological data simulated by WRF as the boundary condition of CFD, which cannot consider the continuous influence of dynamic and random changes in meteorological conditions on pollutant diffusion during simulation. In addition, during the data transfer process, problems such as resolution mismatch, poor terrain adaptability, etc. often lead to information loss and simulation error accumulation, resulting in a large deviation between the predicted results and the actual situation. SUMMARY

[0005] The complex terrain pollutant concentration prediction method based on WRF and CFD coupling of the patent application deeply integrates the macro-weather background of WRF and the micro-terrain effect of CFD, realizes seamless connection simulation from large-scale circulation to small-scale turbulence, significantly improves the spatio-temporal resolution and reliability of pollutant concentration prediction in complex environment, breaks through the precision bottleneck of pollutant prediction in complex terrain, and provides scientific basis for environmental monitoring, risk assessment and pollution prevention and control.

[0006] The complex terrain pollutant concentration prediction method based on WRF and CFD coupling provided by the application comprises the following steps: Step 1: generating a three-dimensional space grid of the simulation area complex terrain according to satellite image elevation data and building model, wherein the simulation area covers the calculation domain of the CFD model; Step 2: generating the evolution of the meteorological field in the first time period of the simulation area by using the WRF model, and extracting the hourly meteorological element data covering the calculation domain of the CFD model, wherein the meteorological element data includes hourly wind speed value, wind direction value, temperature value, air pressure value and humidity value; Step 3: establishing a dynamic coupling mechanism of WRF model and CFD model, and interpolating the grid point data of WRF model grid to the boundary surface grid of CFD model by distance weighting method to obtain boundary conditions and initial field conditions; Step 4: performing numerical calculation based on the CFD model to obtain the flow field data of the pollutant diffusion process under complex terrain conditions.

[0007] Preferably, the step 1 of generating a three-dimensional space grid of the simulation area complex terrain according to satellite image elevation data and building model comprises the following steps: Step 1.1: collecting satellite digital elevation data and building model data, constructing a geometric model, and the geometric model includes three-dimensional complex terrain, buildings and factory model; Step 1.2: loading the geometric model; Step 1.3: setting global and local encryption parameters to automatically generate the three-dimensional space grid of the simulation area.

[0008] Preferably, the spatial resolution of the satellite image elevation data is higher than that of the MODIS data, the spatial resolution of the satellite image elevation data is 20-50 meters, and the spatial resolution of the MODIS data is 900 meters.

[0009] Preferably, the step 2: generating the evolution of the meteorological field in the simulation area in the first time period by using the WRF model, extracting the hourly meteorological element data covering the calculation domain of the CFD model, comprises the following steps: Step 2.1: collecting MODIS data, land use type data, initial meteorological field data of the simulation area; Step 2.2: constructing a multi-layer nested calculation domain of the WRF model according to the range and accuracy requirements of the simulation area; Step 2.3: running the WRF model to simulate the meteorological change process in the simulation area in the first time period, and outputting the hourly meteorological element data of wind speed, wind direction, temperature, air pressure and humidity.

[0010] Preferably, the multi-layer nested calculation domain of the WRF model is a double nested domain, including a WRF outer simulation area and a WRF inner simulation area, the horizontal grid distance of the WRF outer simulation area is 9 km, the horizontal grid distance of the WRF inner simulation area is 3 km, and the range of the WRF inner simulation area covers the calculation domain of the CFD model.

[0011] Preferably, the step 3: establishing a dynamic coupling mechanism of the WRF model and the CFD model, interpolating the grid point data of the WRF model grid to the boundary surface grid of the CFD model by distance weighting method to obtain boundary conditions and initial field conditions, comprises the following steps: Step 3.1: establishing a dynamic coupling mechanism of the WRF model and the CFD model, and transferring the meteorological element data to the CFD model every preset time interval to realize the dynamic interaction of meteorological conditions and flow field; Step 3.2: extracting all grid body center coordinates on the boundary surface of the three-dimensional space grid of the CFD model and setting them as coordinate points to be interpolated; Step 3.3: matching the coordinate points to be interpolated with the grid point coordinates of the WRF model to determine the smallest box of the WRF model containing the coordinate points to be interpolated; Step 3.4: calculating the distance from the position of the coordinate points to be interpolated to each grid point of the smallest box by distance weighting method, and calculating the distance weight coefficient of each grid point of the smallest box; Step 3.5: weighting and averaging the wind speed values of each grid point of the smallest box according to the distance weight coefficient to calculate the wind speed value of the coordinate points to be interpolated, and further obtaining the wind speed boundary condition of the calculation domain of the CFD model; Step 3.6: cyclically traversing the wind speed component data of all WRF model grid points covering the calculation domain of the CFD model, selecting the maximum wind speed value as the reference speed, and establishing the initial field.

[0012] Preferably, the step 3.4: using distance weighting method, calculating the distance from the position of the coordinate point to be interpolated to each grid point of the minimum box, calculating the distance weight coefficient of each grid point of the minimum box, comprises the following steps: Step 3.4.1: using distance weighting method, calculating the distance from the coordinate point to be interpolated to the first i grid point of the minimum box:

[0013] Wherein, d1 is the distance from the coordinate point to be interpolated to the first i grid point of the minimum box, x , y , z is the coordinate value of the coordinate point to be interpolated, is the coordinate value of the first i grid point of the minimum box; Step 3.4.2: calculating the distance weight coefficient of the first i grid point of the minimum box:

[0014] Wherein, is the distance weight coefficient of the first i grid point of the minimum box, N is the number of grid points around the coordinate point to be interpolated to be used in the prediction calculation process; is the exponential coefficient, taking the value of 1-3; with the increase of the distance between the grid point and the coordinate point to be interpolated, the distance weight coefficient value decreases according to the exponential law, and the sum of the distance weight coefficient values of all grid points of the minimum box is 1.

[0015] Preferably, the calculation formula of the wind speed value of the coordinate point to be interpolated in the step 3.5 is:

[0016] Wherein, is the wind speed value calculated at the coordinate point to be interpolated, is the wind speed value of the first i grid point of the minimum box, is the distance weight coefficient of the first i grid point of the minimum box, N is the number of grid points around the coordinate point to be interpolated to be used in the prediction calculation process, x , y , z is the coordinate value of the coordinate point to be interpolated, is the coordinate value of the first i grid point of the minimum box.

[0017] Preferably, the step 4: numerical calculation based on the CFD model, obtaining the flow field data of the pollutant diffusion process under complex terrain conditions, comprising the following steps: Step 4.1: based on the boundary conditions and initial field conditions in step 3, setting the solver parameters and turbulence model; Step 4.2: driving the CFD model software to numerically solve the calculation domain of the CFD model, solving the pollutant transport equation; Step 4.3: after convergence, post-processing is performed to obtain the pollutant concentration distribution at different times and different spatial positions.

[0018] Preferably, the turbulence model can be selected from a Reynolds average model and a large eddy simulation model.

[0019] The complex terrain pollutant concentration prediction method based on WRF and CFD coupling provided in the present application establishes a dynamic coupling mechanism between WRF and CFD, responds to weather changes in real time, and improves dynamic adaptability. In addition, the traditional nonlinear fitting method is abandoned, and a high-precision interpolation method is used to ensure high-fidelity transmission of WRF model coarse resolution meteorological grid point data to CFD model fine resolution grid boundary conditions, avoiding large calculation input errors of the CFD model, and realizing high-precision prediction of complex terrain pollutant concentration. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other embodiments can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0021] Figure 1 is a schematic diagram of a geometric model of a certain mountainous chemical plant.

[0022] Figure 2 is a schematic diagram of a space grid of a certain mountainous chemical plant.

[0023] Figure 3 is a schematic diagram of the interpolation process of WRF model grid point data to CFD model boundary grid.

[0024] Figure 4 is a cloud image schematic diagram of CH4 gas concentration mass fraction distribution in natural gas.

[0025] Figure 5 is a schematic diagram of CH4 gas mass fraction curve comparison at different heights of the centerline position calculated by different methods. DETAILED DESCRIPTION

[0026] In order to better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] Specifically, a complex terrain pollutant concentration prediction method based on WRF and CFD coupling is proposed in the present application, which includes the following steps: Step 1: generating a three-dimensional space grid simulating the complex terrain of the simulation area according to satellite image elevation data and building models.

[0029] Wherein, the complex terrain condition refers to a region with significant spatial heterogeneity of landform, such as mountain (slope ≥ 15°), canyon (width < 500m), plateau, hilly land, etc., and the terrain undulation and roughness change will significantly affect the airflow movement and pollutant diffusion path. Step 1 specifically includes the following steps: Step 1.1: Collect satellite digital elevation (Digital Elevation Model, hereinafter referred to as DEM) data, building model data, and construct three-dimensional complex terrain, building, factory and other geometric models. Wherein, DEM is a digital representation of the earth's surface elevation information obtained by satellite remote sensing and other technologies, which can be used to represent three-dimensional complex terrain, with a spatial resolution of 20 meters to 50 meters, preferably 30 meters. Based on DEM data, a three-dimensional complex terrain geometric model can be constructed by CFD processing software to interpolate discrete elevation data into a continuous three-dimensional terrain surface.

[0030] Step 1.2: Load complex terrain and building, factory and other geometric models. The present embodiment predicts the surrounding pollutant concentration after a high-sulfur natural gas leak at a certain mountainous chemical plant, as shown in Figure 1The figure shows the geometric model of a chemical plant in a mountainous area.

[0031] Step 1.3: Set global and local encryption parameters, automatically generate a spatial grid for the simulation area. As shown in the figure Figure 2 The figure shows the spatial grid of a chemical plant in a mountainous area generated by this embodiment, which is generated from Figure 1 the geometric model of a chemical plant in a mountainous area. The grid is the process of discretizing continuous three-dimensional space into a finite number of small units (such as hexahedron, tetrahedron), which is the basis of CFD calculation. The global encryption parameter is used to set the basic grid size, which can be determined according to the size of the region and the computing resources. When the region is large, a coarser global grid can be used. The local encryption parameter is used to improve the grid resolution in key areas, such as densely populated areas and complex terrain areas.

[0032] Step 2: Use the WRF weather forecast model to generate the evolution of the weather field in the first time period of the WRF simulation area, and extract the weather element data covering the entire CFD mode flow field simulation calculation domain.

[0033] The CFD mode flow field simulation calculation domain is the CFD simulation area, corresponding to the simulation area in step 1. The WRF simulation area covers the simulation area in step 1.

[0034] Step 2.1: Collect Moderate-resolution Imaging Spectroradiometer (MODIS) data, land use type data, and initial weather field data covering the simulation area.

[0035] MODIS is an important remote sensing instrument carried on a satellite, and the data obtained, i.e. MODIS data, is one of the core data sources in the fields of global environmental and climate change monitoring, land cover analysis, and disaster assessment. The WRF model needs terrain height data to accurately represent the influence of surface features on air flow. The MODIS data obtained in this step has a lower spatial resolution than the DEM data in step 1, typically ranging from a few hundred meters to a few kilometers, and is preferably 900 meters. Land use type data is used to define surface properties. The initial weather field data is global weather reanalysis data. The WRF model is a regional model that requires global data as initial and side boundary conditions.

[0036] This embodiment uses low-resolution data simulation to obtain weather data (wind speed, wind direction) as input for calculation, and uses high-resolution CFD simulation to obtain more accurate wind field results.

[0037] Step 2.2: According to the range and accuracy requirements of the simulation area, construct a multi-layer nested calculation domain for the WRF model.

[0038] The embodiment adopts double nesting, including an outer simulation area and an inner simulation area. The outer simulation area is a parent area, covers a larger area, and has a wider grid spacing. The horizontal grid spacing can be 10-15 km, and is preferably 9 km. The inner simulation area is a child area, has a higher resolution, and has a smaller grid spacing. For example, the horizontal grid spacing can be 3-5 km, and is preferably 3 km. The inner simulation area should completely cover the calculation domain of the CFD model.

[0039] Step 2.3: Run the WRF model to simulate the meteorological change process in the simulation area in the first time period, and output hourly meteorological element data such as wind speed, wind direction, temperature, air pressure, and humidity.

[0040] Specifically, the WRF model is used to process high-precision digital DEM data, land use type data, and initial meteorological field data to generate input files such as terrain, land use type, and initial meteorological conditions. The simulation time and output frequency are set to run the WRF model, and hourly meteorological element data such as wind speed, wind direction, temperature, air pressure, and humidity are output.

[0041] The first time period is usually 1 hour to 30 days. For short-term emergency prediction of sudden pollution events, the first time period is usually 1-24 hours. For scientific research-level fine simulation focusing on the influence of terrain on pollutant diffusion, the first time period is usually 3-7 days, which is the most mainstream range. For regional environmental assessment of conventional environmental impact assessment, the first time period is usually 3-30 days. Scenarios exceeding 30 days are rarely seen in actual coupling applications due to insufficient WRF accuracy and high computational cost.

[0042] Step 3: As shown in Figure 3 , a dynamic coupling mechanism between the WRF model and the CFD model is established. The grid point data of the WRF model grid are interpolated to the boundary surface grid of the CFD model by distance weighting method to obtain the boundary conditions and initial field of the calculation domain of the CFD model.

[0043] Step 3.1: A dynamic coupling mechanism between the WRF model and the CFD model is established. The updated meteorological element data of the WRF model are transmitted to the CFD model every certain time interval to realize dynamic interaction between meteorological conditions and flow fields.

[0044] Step 3.2: Extract all grid body center coordinates on the boundary surface of the three-dimensional spatial calculation grid of the CFD model, and set them as coordinate points to be interpolated.

[0045] Step 3.3: Match the coordinate points to be interpolated with the grid point coordinates of the WRF model to determine the smallest box containing the coordinate points to be interpolated.

[0046] Step 3.4: The distance between the coordinate point to be interpolated and each grid point in the minimum box is calculated using the distance weighting method, and the weight of each grid point is calculated.

[0047] The distance between the coordinate point to be interpolated and the first grid point in the minimum box: i (3.1)

[0048] wherein, is the distance between the coordinate point to be interpolated and the first grid point in the minimum box, i , x , y , z is the coordinate value of the coordinate point to be interpolated, is the coordinate value of the first grid point in the minimum box. i In the process of calculating the grid center interpolation on the boundary, the distance weight coefficient of the first grid point in the minimum box is:

[0049] i (3.2) wherein, is the distance weight coefficient of the first grid point in the minimum box,

[0050] is the number of grid points around the coordinate point to be interpolated used in the prediction calculation process; is the exponential coefficient, generally taking a value of 1-3. As the distance between the grid point and the coordinate point to be interpolated increases, the distance weight coefficient value decreases according to the exponential law, and the sum of the distance weight coefficient values of all grid points in the box is 1. i N Step 3.5: The wind speed value of each grid point data in the minimum box is weighted and averaged according to the distance weight coefficient, and the coordinate point to be interpolated is assigned to obtain the boundary condition of the CFD model calculation domain. The calculation formula of the wind speed value calculated at the coordinate point to be interpolated is:

[0051] (3.3)

[0052] wherein, is the wind speed value calculated at the coordinate point to be interpolated, and is the wind speed value of the first grid point.

[0053] i

[0054] ​​​​​The embodiment preferably interpolates wind speed and wind direction data as boundary conditions for the CFD model calculation domain. The wind direction data can be divided into sectors and calculated according to wind frequency, and a general algorithm can be used for the calculation method. Other meteorological data, such as temperature, pressure, and humidity, can be interpolated if the interpolation results are good, or the average value can be taken as the boundary condition if the interpolation results are not good.

[0055] Step 3.6: Loop all wind speed component data of the WRF model grid positions covering the CFD model calculation grid, select the maximum wind speed value as the reference speed, and establish the initial field of the CFD model.

[0056] In step 3, a dynamic coupling mechanism between the WRF model and the CFD model is established, and meteorological data from the WRF model is output to the CFD model. The grid data of the WRF model is interpolated to the boundary surface grid of the CFD model by distance weighting method, and then the CFD model continues to read the meteorological data of the WRF model for calculation, and the process is repeated until the simulation is completed. Thus, the complex terrain pollutant concentration prediction method based on WRF and CFD coupling can realize high-precision and real-time prediction of pollutant concentration under complex terrain conditions by optimizing the data coupling mechanism, improving the data transmission accuracy, and enhancing the dynamic adaptability of meteorological conditions.

[0057] Step 4: Based on the CFD model, numerical calculation is performed to obtain the flow field data of the pollutant diffusion process under complex terrain conditions.

[0058] The embodiment analyzes the influence of complex terrain and buildings on airflow and pollutant diffusion through CFD simulation. CFD can capture micro-scale turbulent effects and terrain details, providing high-precision flow field and pollutant concentration distribution data. Step 4 includes the following steps: Step 4.1: Based on the wind speed boundary and initial field conditions in step 3, set the solver parameters and turbulence model. The turbulence model can be selected from the Reynolds-Averaged Navier-Stokes (RANS) model, Large Eddy Simulation (LES), etc.

[0059] Step 4.2: Drive the CFD model software to numerically solve the CFD model calculation domain, and solve the pollutant transport equation.

[0060] Step 4.3: After the calculation converges, perform post-processing to obtain the pollutant concentration distribution at different times and different spatial positions. After the calculation results are obtained, check the convergence of the calculation results and whether the residual error reaches the preset threshold. If it does, the calculation is complete, and data extraction and statistics can be performed to obtain the pollutant concentration distribution at different times and different spatial positions.

[0061] In this example, the concentration of pollutants in the surrounding area after the leakage of high-sulfur natural gas from a chemical plant in a mountainous area is predicted. Figure 4 This is a schematic diagram of the distribution cloud of the mass fraction of CH4 gas concentration in natural gas during the diffusion process, where the black line is the center line of the leakage. Figure 5 Comparison curves of CH4 gas mass fractions at different heights of the midline position calculated by different methods and measured data are shown. The WRF and CFD coupling method based on distance-weighted interpolation in the present invention improves the simulation accuracy by about 25% compared with the traditional nonlinear function fitting method.

[0062] In summary, the complex terrain pollutant concentration prediction method based on WRF and CFD coupling proposed in the present invention can achieve the following beneficial technical effects: (1) A dynamic coupling mechanism is established between the WRF model and the CFD model, and the simulation process takes into account the impact of changes in meteorological conditions on small-scale flow fields in real time. At regular time steps, the WRF model passes the updated meteorological data to the CFD model as new boundary conditions, solving the problem of static input of meteorological conditions in traditional methods. (2) A non-curve fitting cross-scale coupling technology suitable for the WRF model and the CFD model is adopted, and a distance-weighted interpolation method is used to interpolate the coarse-resolution meteorological data (kilometer level) output by the WRF model to the fine-resolution grid (meter level) required by the CFD model. The interpolation process completely retains the wind field element information of the background grid of the mesoscale model, providing reliable initial and boundary conditions for CFD simulation. Compared with the traditional single-model simulation method, the present invention can more accurately capture the aggregation, diffusion path and concentration distribution characteristics of pollutants in complex terrain, reducing prediction errors.

[0063] It should be pointed out that for ordinary technicians in this field, the technical features in the above embodiments can be freely combined, and the formed technical solutions also belong to the embodiments disclosed in the present invention.

[0064] Furthermore, without departing from the principles of the present invention, several improvements and modifications may be made to the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting pollutant concentration in complex terrain based on WRF and CFD coupling, characterized by: The following steps are involved: Step 1: Generate a 3D spatial grid of the complex terrain of the simulation area based on satellite image elevation data and building models, where the simulation area covers the computational domain of the CFD model; Step 2: Using the WRF model to generate the meteorological field evolution in the first time period of the simulation area, extracting hourly meteorological element data covering the calculation domain of the CFD model, wherein the meteorological element data includes hourly wind speed values, wind direction values, temperature values, air pressure values, and humidity values; Step 3: Establish a dynamic coupling mechanism between the WRF model and the CFD model. Interpolate the grid data of the WRF model grid to the boundary surface grid of the CFD model using the distance weighted method to obtain the boundary conditions and initial field conditions. Step 4: Perform numerical calculations based on the CFD model to obtain flow field data of the pollutant diffusion process under complex terrain conditions.

2. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 1 is characterized in that: The step 1: generating a three-dimensional spatial grid of complex terrain in the simulated area based on satellite image elevation data and building models, includes the following steps: Step 1.1: Collect satellite digital elevation data and building model data to construct a geometric model, which includes three-dimensional complex terrain, buildings, and factory area models; Step 1.2: Load the geometric model; Step 1.3: Set global and local encryption parameters to automatically generate a three-dimensional spatial grid of the simulation area.

3. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 1 is characterized in that: The step 2: using the WRF model to generate the meteorological field evolution in the first time period of the simulation area, and extracting hourly meteorological element data covering the calculation domain of the CFD model, includes the following steps: Step 2.1: Collect MODIS data, land use type data, and initial meteorological field data of the simulation area; Step 2.2: Construct a multi-layer nested computational domain of the WRF model based on the scope and accuracy requirements of the simulation area; Step 2.3: Run the WRF model to simulate the meteorological change process in the first time period of the simulation area, and output hourly meteorological element data of wind speed, wind direction, temperature, air pressure, and humidity.

4. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 3 is characterized in that: The spatial resolution of the satellite image elevation data is higher than that of the MODIS data. The spatial resolution of the satellite image elevation data is 20 meters to 50 meters, while the spatial resolution of the MODIS data is 900 meters.

5. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 3 is characterized in that: The multi-layer nested computational domain of the WRF model is a double nested domain, including a WRF outer simulation area and a WRF inner simulation area. The horizontal grid spacing of the WRF outer simulation area is 9 kilometers, and the horizontal grid spacing of the WRF inner simulation area is 3 kilometers. The range of the WRF inner simulation area covers the computational domain of the CFD model.

6. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 1 is characterized in that: Step 3: establishing a dynamic coupling mechanism between the WRF model and the CFD model, interpolating the grid data of the WRF model grid to the boundary surface grid of the CFD model by the distance weighted method to obtain boundary conditions and initial field conditions, includes the following steps: Step 3.1: Establish a dynamic coupling mechanism between the WRF model and the CFD model. At preset time intervals, the meteorological element data is transmitted to the CFD model to achieve dynamic interaction between meteorological conditions and flow fields. Step 3.2: extracting the coordinates of all mesh body centers on the boundary surface of the three-dimensional space mesh in the CFD model and setting them as coordinate points to be interpolated; Step 3.3: Match the coordinate point to be interpolated with the grid coordinates of the WRF model to determine the smallest box of the WRF model that contains the coordinate point to be interpolated; Step 3.4: Using a distance weighting method, calculate the distance from the position of the coordinate point to be interpolated to each grid point of the minimum box, and calculate the distance weight coefficient of each grid point of the minimum box; Step 3.5: performing a weighted average of the wind speed values ​​of each grid point of the minimum box according to the distance weight coefficient to calculate the wind speed value of the coordinate point to be interpolated, thereby obtaining the wind speed boundary condition of the calculation domain of the CFD model; Step 3.6: Loop through the wind speed component data of all WRF model grid locations covering the computational domain of the CFD model, select the maximum wind speed value as the reference speed, and establish the initial field.

7. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 6 is characterized in that: The step 3.4: using a distance weighting method to calculate the distance from the position of the coordinate point to be interpolated to each grid point of the minimum box, and calculating the distance weight coefficient of each grid point of the minimum box, comprises the following steps: Step 3.4.1: Use the distance weighting method to calculate the distance from the coordinate point to be interpolated to the minimum box. i The distance between grid points: in, The first value from the coordinate point to be interpolated to the minimum box i The distance between grid points, x 、 y 、 z is the coordinate value of the coordinate point to be interpolated, 、 、 is the smallest box i The coordinate values ​​of the grid points; Step 3.4.2: Calculate the minimum box i The distance weight coefficient of each grid point: in, is the smallest box i The distance weight coefficient of each grid point, N The number of grid points around the coordinate point to be interpolated to be used in the prediction calculation process; is an exponential coefficient with a value of 1 to 3. As the distance between the grid point and the coordinate point to be interpolated increases, the distance weight coefficient value decreases according to the exponential law, and the sum of the distance weight coefficient values ​​of all grid points in the minimum box is 1.

8. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 6 is characterized in that: The calculation formula for the wind speed value of the coordinate point to be interpolated in step 3.5 is: in, is the wind speed value calculated at the coordinate point to be interpolated, and U is the wind speed value of the smallest box. i The wind speed value at each grid point, is the smallest box i The distance weight coefficient of each grid point, N To predict the number of grid points around the coordinate point to be interpolated to be used in the calculation process, x 、 y 、 z is the coordinate value of the coordinate point to be interpolated, is the smallest box i The coordinate values ​​of the grid points.

9. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 1 is characterized in that: Step 4: performing numerical calculations based on the CFD model to obtain flow field data of the pollutant diffusion process under complex terrain conditions, including the following steps: Step 4.1: Based on the boundary conditions and initial field conditions in step 3, set the solver parameters and turbulence model; Step 4.2: driving the CFD model software to numerically solve the computational domain of the CFD model to solve the pollutant transport equation; Step 4.3: After the calculation converges, post-processing is performed to obtain the pollutant concentration distribution at different times and different spatial locations.

10. The complex terrain pollutant concentration prediction method based on WRF and CFD coupling according to claim 9 is characterized in that: The turbulence model may be a Reynolds average model or a large eddy simulation model.

Images