Multi-scale coupling solving method suitable for wind field simulation

Abstract

The invention relates to the field of wind engineering, and provides a multi-scale coupling solving method suitable for wind field simulation. The method aims at solving the problems that in an existing wind field simulation technology, due to the fact that wind speed fluctuation from the ground to the high altitude is large and irregular due to the random characteristics of a wind field, a curve fitting function is difficult to accurately simulate the change rule of the wind speed fluctuation, large deviation is generated, and the calculation accuracy is seriously affected. According to the method, the mesoscale weather forecast WRF mode is coupled with the CFD mode, the distance weighting method is used for interpolating the grid point data of the WRF mode to the boundary surface grid of the CFD mode to serve as the boundary condition and the initial field, and the problem of input condition errors caused by nonlinear fitting in traditional simulation is effectively solved. By means of the method, high-fidelity data transmission is achieved, and the accuracy and reliability of wind field simulation under the complex terrain condition are remarkably improved. The method has a wide application prospect and can be applied to multiple fields such as wind energy resource assessment, wind disaster early warning and urban planning.

Description

Technical Field

[0001] The invention relates to the field of wind engineering and provides a multi-scale coupling solution method suitable for wind field simulation. Background Art

[0002] As an inexhaustible renewable energy source, the use of wind energy has had a profound impact on the development of human society. However, frequent and strong wind disasters often bring serious damage, casualties and huge economic losses. In addition, complex terrain also leads to irregular distribution of characteristics such as near-ground wind speed and wind pressure, further exacerbating the difficulty of wind field prediction. Therefore, it is particularly important to achieve accurate prediction of wind fields.

[0003] In the field of mesoscale meteorological forecasting, models such as WRF, MM5, and RAMS are widely used. Among them, the WRF model has become a leader in this field with its advanced numerical methods, data assimilation technology, and optimized physical process schemes. This model has multiple nesting capabilities and can be easily located in different geographical locations. It focuses on horizontal grids with a resolution of several kilometers to tens of kilometers, so as to accurately simulate complex weather phenomena, and is especially suitable for large-area wind field simulation. However, due to the limitations of its grid resolution and the influence of the choice of physical schemes, the mesoscale model is difficult to capture small-scale terrain environmental changes, especially in complex terrain areas, and the relative error of wind field simulation is large.

[0004] In contrast, microscale models have advantages in simulating wind fields in smaller-scale complex terrains. Among them, the model based on computational fluid dynamics (CFD) discretizes the continuous physical quantity fields in the time domain and space domain, and then uses numerical calculations to solve the basic equations of fluid motion (such as mass, momentum, energy equations, etc.), thereby obtaining the basic physical quantities at each position in the flow field and their changing relationship over time. CFD numerical calculation has the characteristics of fast calculation speed and high efficiency, and can achieve refined simulation of wind fields. However, since the CFD model is affected by idealized boundary conditions, there is a certain gap between its simulation results and the actual wind field, resulting in certain calculation errors.

[0005] In recent years, researchers have tried to achieve downscaling simulation of wind fields through coupling calculations of mesoscale models and CFD models (such as Fluent, Windsim and other commercial software) to improve the accuracy of near-surface wind field simulations in complex terrain. However, existing methods usually use curve fitting functions for calculations. Due to the random characteristics of the wind field, the wind speed fluctuations from the ground to high altitudes are large and have no obvious regularity, making it difficult for the curve fitting function to fully simulate its variation pattern, which is prone to large deviations, seriously affecting the calculation accuracy of wind field simulations. Summary of the invention

[0006] The purpose of the present invention is to solve the problem in the existing wind field simulation method that the wind speed fluctuation from the ground to the high altitude is large and has no obvious regularity due to the random characteristics of the wind field, making it difficult for the curve fitting function to fully simulate its variation law, and thus prone to large deviations, which seriously affects the calculation accuracy. To this end, the present application proposes a multi-scale coupling solution method suitable for wind field simulation, which effectively solves the problem of downscaling the background wind field information elements from sparse grids to dense grids, thereby improving the accuracy and reliability of wind field simulation.

[0007] In order to achieve the above purpose, the present invention adopts the following technical means:

[0008] A multi-scale coupling solution method suitable for wind field simulation includes the following steps:

[0009] Step 1: Generate a three-dimensional space computational grid for the CFD model based on the geometric model of the target area;

[0010] Step 2: Extract wind field information elements covering the entire CFD model flow field simulation calculation domain in the mesoscale meteorological forecast WRF model, and obtain multi-layer wind speed component forecast data at the sample node position of the three-dimensional space calculation grid; (In fact, multi-layer wind speed component forecast data is actually wind speed data at different heights, and the WRF model output data is layered output in the height direction)

[0011] Step 3: Interpolate the WRF model grid data to the boundary surface grid of the CFD model through the distance weighted method to obtain the boundary conditions and initial fields;

[0012] Step 4: Drive the CFD model to perform numerical calculations to obtain refined wind field information data.

[0013] In the above scheme, step 1 specifically includes the following steps:

[0014] Step 1.1: Import the geometric model of the target area into the mesh generation software for meshing;

[0015] Step 1.2: Output the divided three-dimensional space grid in the required format to generate the three-dimensional space calculation grid of the CFD model software.

[0016] In the above scheme, step 2 specifically includes the following steps:

[0017] Step 2.1: Establish a mesoscale WRF forecast model for the target area, select the type of meteorological data source, and obtain meteorological data information for the forecast period;

[0018] Step 2.2: Based on the meteorological data of the WRF forecast model, identify the location information of the grid nodes in the target area;

[0019] Step 2.3: According to the grid node position information, extract the multi-layer wind speed component forecast data at the sample node position of the three-dimensional space calculation grid covering the CFD model.

[0020] In the above scheme, step 3 specifically includes the following steps:

[0021] Step 3.1: Extract the body center coordinates of all boundary surface mesh units of the boundary surface of the three-dimensional space calculation mesh of the CFD model and set them as the coordinate points to be interpolated;

[0022] Step 3.2: Match the coordinate points to be interpolated with the sample node coordinates of the WRF mode to determine the minimum box of the sample node that contains the coordinate points to be interpolated;

[0023] Step 3.3: Use the distance weighting method to calculate the minimum box from the center coordinate position of the boundary surface grid unit to the sample node where it is located ( Figure 3 This step is explained in the figure, that is, the point to be interpolated must be within the box formed by the sample nodes in the WRF mode, and then these sample points are used for interpolation. ) The distance is calculated from this, and the weight contributed by each sample node is calculated;

[0024] The distance from the body center coordinates of the boundary surface mesh unit to the i-th sample node of the box:

[0025]

[0026] Among them, x, y, and z are the body-centered coordinates of the boundary surface grid unit to be interpolated on the boundary, x i ,y i 、z i is the coordinate value of the ,th sample node of the box;

[0027] During the calculation of the body-centered coordinate interpolation of the boundary surface mesh unit, the distance weight coefficient of the th sample node of the box is:

[0028]

[0029] N is the number of sample nodes around the interpolation point to be used in the calculation process;

[0030] P is the exponential coefficient and its value is between 1 and 3;

[0031] As the distance between the sample node and the point to be interpolated increases, the weight coefficient value decreases exponentially, and the sum of the weight coefficient values ​​is 1;

[0032] Step 3.4: Perform weighted average calculation on the wind speed value of each sample node according to the weight coefficient of the data, assign the boundary surface value grid unit center velocity, and then obtain the boundary input conditions of the CFD model calculation domain;

[0033] The wind speed calculated at the center of the boundary surface mesh element is:

[0034]

[0035] U CFD (x, y, z) represents the velocity of the mesh cell center at the boundary surface in the CFD model, U(x i ,y i , z i ) represents the velocity of the body-center coordinates of the mesh unit on the boundary surface to be interpolated at the i-th sample node in the box, λ: is the weight coefficient of the i-th sample node, and N represents the number of sample nodes.

[0036] Step 3.5: Loop through all the multi-layer wind speed component data at the sample node positions covering the CFD model calculation grid, select the maximum wind speed value as the reference speed, and establish the initial field.

[0037] In the above scheme, step 4 specifically includes the following steps:

[0038] Step 4.1: Based on the above boundary conditions and initial field conditions, set the solver parameters and turbulence model;

[0039] Step 4.2: Drive the CFD model software to perform numerical calculations on the calculation domain;

[0040] Step 4.3: After the calculation converges, post-processing analysis is performed to obtain high-resolution flow field distribution data.

[0041] Because the present invention adopts the above technical means, it has the following beneficial effects:

[0042] 1. The multi-scale coupling solution method suitable for wind field simulation proposed in this application abandons the traditional nonlinear fitting method, avoids the input condition errors caused by it, ensures the high-fidelity transmission of the grid data of the mesoscale weather forecast model through high-precision interpolation methods, and improves the simulation capability of wind fields under complex terrain conditions.

[0043] 2. The present invention uses the inverse distance weighted method to interpolate WRF mesoscale data to the CFD calculation grid as boundary conditions and initial field conditions, which effectively overcomes the input condition error problem caused by nonlinear fitting in traditional simulations, realizes high-fidelity data transmission, and significantly improves the accuracy and reliability of wind field simulation under complex terrain conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a simplified flow chart of the present invention;

[0045] Figure 2 This is a simplified flow chart of the implementation of step 3 of the present invention;

[0046] Figure 3 Three-dimensional CFD computational grid division in a mountainous area;

[0047] Figure 4 Data transfer from mesoscale model to CFD model;

[0048] Figure 5 The interpolation process of sample node data to boundary grid;

[0049] Figure 6 The wind speeds predicted and measured at a height of 100 meters above the ground in a mountainous area were compared using different methods, among which the multi-scale coupling method was the method of the present invention. DETAILED DESCRIPTION

[0050] The following is a detailed description of the embodiments of the present invention. Although the present invention will be described and illustrated in conjunction with some specific embodiments, it should be noted that the present invention is not limited to these embodiments. On the contrary, modifications or equivalent substitutions made to the present invention should all be included in the scope of the claims of the present invention.

[0051] In addition, in order to better illustrate the present invention, numerous specific details are given in the following specific embodiments. It will be understood by those skilled in the art that the present invention can also be implemented without these specific details.

[0052] See also Figure 4 The comparison curves of wind speed predicted and measured using different methods at a height of 100m above the ground in a mountainous area show that the calculation results of the multi-scale coupling solution method are closer to the measured wind speed results than the traditional nonlinear fitting method. The average error of the wind speed calculation is 0.7446m / s, the calculation accuracy is 84.61%, and the corresponding correlation coefficient is as high as 0.9614.

[0053] The present invention provides a multi-scale coupling solution method suitable for wind field simulation, comprising the following steps:

[0054] Step 1: Generate a three-dimensional space computational grid for the CFD model based on the geometric model of the target area;

[0055] Step 2: Extract wind field information elements covering the entire CFD model flow field simulation calculation domain in the mesoscale meteorological forecast WRF model, and obtain multi-layer wind speed component forecast data at the sample node position of the three-dimensional space calculation grid; (In fact, the multi-layer wind speed component forecast data is actually the wind speed data at different heights, and the WRF model output data is layered output in the height direction).

[0056] Step 3: Interpolate the WRF model grid data to the boundary surface grid of the CFD model through the distance weighted method to obtain the boundary conditions and initial fields;

[0057] Step 4: Drive the CFD model to perform numerical calculations to obtain refined wind field information data.

[0058] In the above scheme, step 1 specifically includes the following steps:

[0059] Step 1.1: Import the geometric model of the target area into the mesh generation software for meshing;

[0060] Step 1.2: Output the divided three-dimensional space grid in the required format to generate the three-dimensional space calculation grid of the CFD model software.

[0061] In the above scheme, step 2 specifically includes the following steps:

[0062] Step 2.1: Establish a mesoscale WRF forecast model for the target area, select the type of meteorological data source, and obtain meteorological data information for the forecast period;

[0063] Step 2.2: Based on the meteorological data of the WRF forecast model, identify the location information of the grid nodes in the target area;

[0064] Step 2.3: According to the grid node position information, extract the multi-layer wind speed component forecast data at the sample node position of the three-dimensional space calculation grid covering the CFD model.

[0065] In the above scheme, step 3 specifically includes the following steps:

[0066] Step 3.1: Extract the body center coordinates of all boundary surface mesh units of the boundary surface of the three-dimensional space calculation mesh of the CFD model and set them as the coordinate points to be interpolated;

[0067] Step 3.2: Match the coordinate points to be interpolated with the sample node coordinates of the WRF mode to determine the minimum box of the sample node that contains the coordinate points to be interpolated;

[0068] Step 3.3: Use the distance weighting method to calculate the minimum box from the center coordinate position of the boundary surface grid unit to the sample node where it is located ( Figure 3 This step is explained in the figure, that is, the point to be interpolated must be within the box formed by the sample nodes in the WRF mode, and then these sample points are used for interpolation. ) The distance is calculated from this, and the weight contributed by each sample node is calculated;

[0069] The distance from the body center coordinates of the boundary surface mesh unit to the i-th sample node of the box:

[0070]

[0071] Among them, x, y, and z are the body-centered coordinates of the boundary surface grid unit to be interpolated on the boundary, x i,y i 、z i is the coordinate value of the i-th sample node of the box;

[0072] During the calculation of the body-centered coordinate interpolation of the boundary surface mesh unit, the distance weight coefficient of the th sample node of the box is:

[0073]

[0074] N is the number of sample nodes around the interpolation point to be used in the calculation process;

[0075] P is the exponential coefficient and its value is between 1 and 3;

[0076] As the distance between the sample node and the point to be interpolated increases, the weight coefficient value decreases exponentially, and the sum of the weight coefficient values ​​is 1;

[0077] Step 3.4: Perform weighted average calculation on the wind speed value of each sample node according to the weight coefficient of the data, assign the boundary surface value grid unit center velocity, and then obtain the boundary input conditions of the CFD model calculation domain;

[0078] The wind speed calculated at the center of the boundary surface mesh element is:

[0079]

[0080] U CFD (x, y, z) represents the velocity of the mesh cell center at the boundary surface in the CFD model, U(x i ,y i , z i ) represents the velocity of the i-th sample node in the box at the center coordinate of the mesh unit on the boundary surface to be interpolated, λ i is the weight coefficient of the i-th sample node, and N represents the number of sample nodes.

[0081] Step 3.5: Loop through all the multi-layer wind speed component data at the sample node positions covering the CFD model calculation grid, select the maximum wind speed value as the reference speed, and establish the initial field.

[0082] In the above scheme, step 4 specifically includes the following steps:

[0083] Step 4.1: Based on the above boundary conditions and initial field conditions, set the solver parameters and turbulence model;

[0084] Step 4.2: Drive the CFD model software to perform numerical calculations on the calculation domain;

[0085] Step 4.3: After the calculation converges, post-processing analysis is performed to obtain high-resolution flow field distribution data.

[0086] The multi-scale coupling solution method for wind field simulation proposed in the application is a WRF forecast model data and CFD model coupling solution technology, which uses a distance weighted interpolation method to accurately input the grid data of the mesoscale WRF forecast model into the microscale CFD model. The mesoscale WRF forecast model grid data is directly interpolated to the center of the boundary surface grid in the microscale CFD model through a custom function interface. This process completely retains the background grid wind field element information of the mesoscale model, thereby effectively improving the calculation accuracy of wind field simulation.

[0087] The multi-scale coupling solution method suitable for wind field simulation proposed in this application abandons the traditional nonlinear fitting method, avoids the input condition errors caused by it, ensures the high-fidelity transmission of the grid data of the mesoscale weather forecast model through high-precision interpolation methods, and improves the simulation capabilities of wind fields under complex terrain conditions.

Claims

1. A multi-scale coupling solution method suitable for wind field simulation, characterized in that: The following steps are involved: Step 1: Generate a three-dimensional space computational grid for the CFD model based on the geometric model of the target area; Step 2: Extract wind field information elements covering the entire CFD model flow field simulation calculation domain in the mesoscale meteorological forecast WRF model, and obtain multi-layer wind speed component forecast data at the sample node position of the three-dimensional space calculation grid; Step 3: Interpolate the WRF model grid data to the boundary surface grid of the CFD model through the distance weighted method to obtain the boundary conditions and initial fields; Step 4: Drive the CFD model to perform numerical calculations to obtain refined wind field information data.

2. A multi-scale coupling solution method suitable for wind field simulation according to claim 1, characterized in that: Step 1 specifically includes the following steps: Step 1.1: Import the geometric model of the target area into the mesh generation software for meshing; Step 1.2: Output the divided three-dimensional space grid in the required format to generate the three-dimensional space calculation grid of the CFD model software.

3. A multi-scale coupling solution method suitable for wind field simulation according to claim 1, characterized in that: Step 2 specifically includes the following steps: Step 2.1: Establish a mesoscale WRF forecast model for the target area, select the type of meteorological data source, and obtain meteorological data information for the forecast period; Step 2.2: Based on the meteorological data of the WRF forecast model, identify the location information of the grid nodes in the target area; Step 2.3: According to the grid node position information, extract the multi-layer wind speed component forecast data at the sample node position of the three-dimensional space calculation grid covering the CFD model.

4. A multi-scale coupling solution method suitable for wind field simulation according to claim 1, characterized in that: Step 3 specifically includes the following steps: Step 3.1: Extract the body center coordinates of all boundary surface mesh units of the boundary surface of the three-dimensional space calculation mesh of the CFD model and set them as the coordinate points to be interpolated; Step 3.2: Match the coordinate points to be interpolated with the sample node coordinates of the WRF mode to determine the minimum box of the sample node that contains the coordinate points to be interpolated; Step 3.3: Use the distance weighting method to calculate the distance from the center coordinate position of the boundary surface grid unit to the minimum box of the sample node where it is located, and calculate the weight of each sample node's contribution from this; The distance from the body center coordinates of the boundary surface mesh unit to the i-th sample node of the box: Among them, x, y, and z are the body-centered coordinates of the boundary surface grid unit to be interpolated on the boundary, x i ,y i 、z i is the coordinate value of the i-th sample node of the box; During the calculation of the body-centered coordinate interpolation of the boundary surface mesh unit, the distance weight coefficient of the i-th sample node of the box is: N is the number of sample nodes around the interpolation point to be used in the calculation process; P is the exponential coefficient and its value is between 1 and 3; As the distance between the sample node and the point to be interpolated increases, the weight coefficient value decreases exponentially, and the sum of the weight coefficient values ​​is 1; Step 3.4: Perform weighted average calculation on the wind speed value of each sample node according to the weight coefficient of the data, assign the boundary surface value grid unit center velocity, and then obtain the boundary input conditions of the CFD model calculation domain; The wind speed calculated at the center of the boundary surface mesh element is: U CFD (x, y, z) represents the velocity of the mesh cell center on the boundary surface in the CFD model, U(x i ,y i , z i ) represents the velocity of the i-th sample node in the box at the center coordinate of the mesh unit on the boundary surface to be interpolated, λ i is the weight coefficient of the i-th sample node, and N represents the number of sample nodes. Step 3.5: Loop through all the multi-layer wind speed component data at the sample node positions covering the CFD model calculation grid, select the maximum wind speed value as the reference speed, and establish the initial field.

5. The multi-scale coupling solution method suitable for wind field simulation according to claim 1, characterized in that: Step 4 specifically includes the following steps: Step 4.1: Based on the above boundary conditions and initial field conditions, set the solver parameters and turbulence model; Step 4.2: Drive the CFD model software to perform numerical calculations on the calculation domain; Step 4.3: After the calculation converges, post-processing analysis is performed to obtain high-resolution flow field distribution data.

Images