A real-time assimilation method for multi-scale numerical simulation of wind field of complex terrain by fusing external measurement data

Abstract

The application discloses a complex terrain wind field multi-scale numerical simulation real-time assimilation method fusing external field measurement data, and comprises the following steps: acquiring conventional meteorological and terrain data, determining a reference point, simulating complex terrain wind field numbers based on a mesoscale WRF mode; acquiring external field observation data and high-precision terrain data, combining high-precision terrain and high-resolution grids, and improving the mesoscale numerical simulation precision of the complex terrain wind field to meet a first preset precision requirement based on a four-dimensional variation assimilation method; establishing a meso-micro scale coupling simulation framework based on WRF-CFD mode coupling, and performing high-fidelity CFD numerical simulation on a target wind field; and applying the four-dimensional variation assimilation technology to the CFD simulation process to improve the CFD numerical simulation precision to meet a second preset precision requirement. The application can improve the multi-scale numerical simulation precision of the complex terrain wind field.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wind power generation technology, in particular to a complex terrain wind field multi-scale numerical simulation real-time assimilation method fusing external field measurement data. BACKGROUND

[0002] With the rapid development of global wind power industry, wind power generation has become one of the key technologies to achieve energy transformation and respond to climate change. However, the wind field in complex terrain has the characteristics of variable wind direction, high turbulence intensity, unsteady and multi-scale, which brings great challenges to the accurate assessment of wind energy resources. The traditional mesoscale-microscale coupling simulation method, such as the combination of WRF model and CFD technology, can take into account the mesoscale atmospheric conditions and microscale terrain effects, but still has deficiencies in simulation accuracy.

[0003] In mesoscale simulation, WRF model can simulate the temporal and spatial variation of wind speed and turbulence, but its accuracy is limited by the assumptions of planetary boundary layer (PBL) model, large grid spacing and insufficient terrain resolution. The data assimilation method, especially the Observational Nudging method based on four-dimensional variational assimilation, can improve the accuracy of mesoscale WRF simulation, but when applied to complex terrain wind field simulation, it is affected by grid spacing, terrain resolution and assimilation frequency, resulting in loss of key terrain flow details, distortion of wind field simulation, and limitation of overall simulation accuracy improvement.

[0004] In microscale simulation, the RANS model based on the eddy viscosity hypothesis has problems such as closure model error and inaccurate turbulence generation, which limits the simulation fidelity. The data assimilation method based on LES has shown potential to improve simulation results in microscale CFD simulation, but the application of Observational Nudging method to complex terrain wind field microscale CFD simulation still faces many challenges, such as optimization of time parameter, definition of horizontal and vertical spatial weight function, determination of horizontal and vertical influence range, etc. are not clear.

[0005] In summary, there is a lack of a complex terrain wind field multi-scale simulation technology that can integrate data assimilation method into WRF process and microscale CFD simulation, and fully consider factors such as grid resolution, terrain resolution, assimilation frequency and unsteady effect. Therefore, it is of great practical significance and application value to develop a new meso-microscale coupling simulation technology based on data assimilation method to improve the numerical simulation accuracy of complex terrain wind field. SUMMARY

[0006] The application aims to provide a complex terrain wind field multi-scale numerical simulation real-time assimilation method fusing external field measurement data.

[0007] To achieve the above technical purposes, the technical scheme adopted by the application is:

[0008] A complex terrain wind field multi-scale numerical simulation real-time assimilation method fusing external field measurement data, the method comprising the following steps:

[0009] S1, obtaining conventional meteorological and terrain data, determining a reference point, and simulating a complex terrain wind field value based on a mesoscale WRF model; judging the complex terrain wind field mesoscale numerical simulation accuracy, if the first preset accuracy requirement is met, turning to step S3, otherwise, turning to step S2;

[0010] S2, obtaining external field observation data and high-precision terrain data, combining high-precision terrain and high-resolution grid, and improving the complex terrain wind field mesoscale numerical simulation accuracy to meet the first preset accuracy requirement based on a four-dimensional variational assimilation method;

[0011] S3, establishing a meso-micro scale coupling simulation framework based on WRF-CFD model coupling, and performing high-fidelity CFD numerical simulation on the target wind field; judging the CFD numerical simulation accuracy, if the second preset accuracy requirement is met, ending the process, otherwise, turning to step S4;

[0012] S4, applying four-dimensional variational assimilation technology to the CFD simulation process to improve the CFD numerical simulation accuracy to meet the second preset accuracy requirement.

[0013] Step S1 further comprises the following steps:

[0014] S11, determining conventional climate input data and terrain data;

[0015] S12, selecting a wind field center as the reference point of the wind field to be simulated, obtaining its longitude and latitude; removing the start time of the mesoscale WRF model, and determining the simulation time period of the wind field to be simulated;

[0016] S13, determining the calculation domain range, grid scale, nesting mode, parameterization scheme configuration, monitoring point position and monitoring variable of the mesoscale WRF simulation;

[0017] S14, performing complex terrain wind field mesoscale numerical simulation and result analysis.

[0018] Further, in step S13, the calculation domain range is determined according to the reference point position and the target wind field size; the grid scale and the time scale of each nested domain are determined according to the underlying surface terrain complexity; the nested mode is selected as a bidirectional nesting; the parameterization scheme includes microphysical processes, shortwave and longwave radiation, surface and land surface processes, urban and ocean models, planetary boundary layer schemes and cumulus parameterization processes; the monitoring position is the wind tower position, and the monitoring variables include wind speed, pressure and potential temperature.

[0019] Further, in step S1, the process of judging the mesoscale numerical simulation accuracy of the complex terrain wind field includes:

[0020] Obtaining the external field observation data, the external field observation data including the longitude, latitude and altitude of the observation site and the wind speed, wind direction, temperature, humidity and air pressure variables of each observation site; and using the observation data to verify whether the mesoscale numerical simulation accuracy of the complex terrain wind field meets the first preset accuracy requirement.

[0021] Further, in step S2, the four-dimensional variational assimilation method is used to improve the mesoscale numerical simulation accuracy of the complex terrain wind field, and the data assimilation expression is:

[0022] ;

[0023] wherein i is the index of the observation data, x, y and z are the spatial dimension coordinates, t represents the simulation time, q represents the physical quantity to be assimilated, μ is the dry air static pressure, q0 is the observation value of the physical quantity, q m (x i ,y i ,z i ,t) is the prediction value of the WRF model, G q is the assimilation intensity of q, N is the total number of observation data, W q is the weighting function based on the distance between the grid cell and the observation point; W t , W z and W xy describe the assimilation intensity of time and space; W xy and W z are the assimilation spatial weight functions, and W t is the time weight function.

[0024] ;

[0025] ;

[0026] In the formula, R is the maximum horizontal influence radius, D is the horizontal distance between the grid point and the observation point, t0 is the observation time point, t is the simulation time, is the length of the time window of the observation data; W t The value range of the parameter is 0 to 1.

[0027] Step S3 further comprises:

[0028] S31, a meso-micro scale coupling simulation framework based on WRF-CFD model coupling is established, and boundary coupling method is used for WRF-CFD data transmission; specifically, the wind speed, temperature and turbulent kinetic energy parameter space-time sequence corresponding to the boundary of the micro scale calculation domain is extracted from the WRF simulation result, and the extracted data is imported into the boundary of the CFD simulation calculation domain through time and space interpolation for micro scale calculation;

[0029] S32, high-fidelity CFD numerical simulation is performed on the target wind field;

[0030] S33, the WRF-CFD numerical simulation result is verified based on the field measured data.

[0031] In step S31, the establishment process of the coupling boundary comprises the following steps:

[0032] S311, the longitude and latitude and coordinates of the reference point are determined, and the reference point is located at the center point of the mesoscale calculation domain and the micro scale calculation domain;

[0033] S312, the boundary of the micro scale calculation domain is determined according to the height of the mountain and the thickness of the atmospheric boundary layer;

[0034] S313, the boundary coordinates of the micro scale calculation domain are extracted;

[0035] S314, the wind speed, turbulent kinetic energy, temperature and pressure information corresponding to the boundary coordinates are extracted from the mesoscale numerical simulation result, and the extracted boundary variables are respectively subjected to spatial interpolation and time interpolation, the interpolated boundary variables meet the micro scale grid requirement in space and meet the micro scale simulation time step requirement in time;

[0036] S315, the exported boundary variables are added to the boundary of the micro scale calculation domain to complete the CFD numerical simulation,

[0037] Step S4 further comprises:

[0038] S41, a source term F and a modified velocity field U are added in the momentum equation, and the momentum equation expression is modified as:

[0039] ;

[0040] In the formula: is the velocity component, and the subscript respectively represents the flow direction, the transverse direction and the vertical direction, is the coordinate component, is the time, is the fluid density, is the static pressure, is the potential temperature, is the horizontal average; is the gravity acceleration, is the reference potential temperature, is the Coriolis force coefficient, is the average angular velocity of the earth rotation, is the local latitude, , represent the streamwise and crosswise components of the geostrophic wind respectively; is the Dirac function, j = 1, 2, 3, when , , otherwise 0; is the stress, is the source term for velocity correction.

[0041] S42, adjust the flow field velocity to the target wind speed profile by updating the source term F.

[0042] Step S42 further comprises:

[0043] S421, calculate the weighted average wind speed in the grid volume specified by the target position:

[0044] ;

[0045] wherein, represents the wind speed vector in the i-th grid cell corresponding to the j-th assimilation point, represents the volume in the i-th grid cell corresponding to the j-th assimilation point, j is a certain point that needs to be assimilated, and m is the number of grids in the specified volume; after removing the vertical component, we get:

[0046] ;

[0047] wherein, represents the unit normal vector in the vertical direction;

[0048] S422, read the wind speed vector of the target position at the nearest time point from the wind speed time series ;

[0049] S423, based on the difference between the target wind speed and the current average wind speed, calculate the wind speed adjustment amount ds and normalize it:

[0050] ;

[0051] ;

[0052] ;

[0053] where α is the relaxation factor, is the volume-weighted average of the inverse of the diagonal matrix of the momentum equation rAU;

[0054] S424, the spatial and temporal weight functions are calculated; the horizontal weight is:

[0055] ;

[0056] where r xy is the maximum horizontal influence radius, d xy is the horizontal distance between the grid point and the observation point;

[0057] The vertical weight is:

[0058] ;

[0059] where r z is the maximum vertical influence range, d z is the vertical distance between the grid point and the observation point;

[0060] The temporal weight is:

[0061] ;

[0062] where t0 is the observation time point, t is the simulation time, is the time window length of the observation data;

[0063] The total weight is: ;

[0064] where , , respectively represent the assimilation spatial weight function and the temporal weight function of the jth assimilation point;

[0065] S425, the adjustment amount is distributed to the grid according to the weight, and the F and the velocity field U are updated as:

[0066] ;

[0067] ;

[0068] The flux field φ is updated synchronously as:

[0069] ;

[0070] where n is the number of grids in the calculation domain, face interpolation field of the inverse of the diagonal matrix of the momentum equation, rAU, used to build the pressure correction equation to guarantee the numerical stability of the velocity-pressure coupling, fvc denotes the finite volume convection, interpolate denotes the interpolation function used to interpolate the physical quantities at the cell centers to the grid interfaces, denotes the area vector of the grid interface.

[0071] Compared with the prior art, the present application has the following beneficial effects:

[0072] It is found through comparison with the measured data of the wind field that the wind speed, wind direction and turbulence intensity obtained by the complex terrain wind field multi-scale numerical simulation real-time assimilation method fusing external field measurement data of the present application have good consistency with the observation data. The data assimilation method has the advantages of simple form, easy coding, high calculation efficiency and high calculation precision, and has good prediction accuracy for the spatial and temporal distribution of wind speed, wind direction and turbulence. The calculation accuracy is better than the numerical simulation results based on the coupling method of WRF model and WRF-CFD model. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1a is the application process schematic diagram of four-dimensional variational assimilation technology (Observational Nudging) in the mesoscale WRF model;

[0074] Figure 1b is the horizontal weight function calculation schematic diagram of four-dimensional variational assimilation technology applied to WRF model;

[0075] Figure 1c is the time weight function calculation schematic diagram of four-dimensional variational assimilation technology applied to WRF model;

[0076] Figure 2 is the establishment process schematic diagram of coupling boundary in the coupling framework of WRF-CFD model;

[0077] Figure 3 is the application process schematic diagram of four-dimensional variational assimilation technology (Observational Nudging) in the microscale CFD calculation process;

[0078] Figure 4 is the topographic schematic diagram of a certain wind field radar position (the radar is used to obtain external field measured data);

[0079] Figure 5a is the schematic diagram of conventional terrain static data (USGS, based on mesoscale WRF model to simulate complex terrain wind field);

[0080] Figure 5b is the schematic diagram of high-resolution terrain static data (SRTM 30, based on mesoscale WRF model to simulate complex terrain wind field);

[0081] Figure 6a Fig. 1 is a schematic diagram of a micro-scale CFD calculation domain and grid when simulating a complex terrain wind field based on a WRF-CFD model coupling framework;

[0082] Figure 6b Fig. 2 is a schematic diagram of a complex terrain wind field simulation result based on a WRF-CFD model coupling framework; Figure 6a Fig. 3 is a partial enlarged view of the graph in the middle red rectangular frame;

[0083] Figure 7a Fig. 4 is a schematic diagram of the change of wind speed and direction with time at a certain wind field radar position (first row: wind speed; second row: wind direction; wind speed and direction at a radar position 101 m high, 2022.12.22 0:00-24:00, UTC+8, local time);

[0084] Figure 7b Fig. 5 is a schematic diagram of the change of wind speed and direction with height at a certain wind field radar position (left: wind speed; right: wind direction; 2022.12.22 15:00-16:00, UTC+8, local time);

[0085] Figure 7c Fig. 6 is a schematic diagram of the change of turbulence intensity with height at a certain wind field radar position (2022.12.22 15:00-16:00, UTC+8, local time);

[0086] Figure 8 Fig. 7 is a schematic diagram of a real-time assimilation method for multi-scale numerical simulation of complex terrain wind field fusing external measurement data according to the present application. DETAILED DESCRIPTION

[0087] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0088] Referring to Figure 8 , the embodiments of the present application disclose a real-time assimilation method for multi-scale numerical simulation of complex terrain wind field fusing external measurement data, comprising:

[0089] Step 1, obtaining conventional meteorological and terrain data, determining a reference point, and numerically simulating a complex terrain wind field based on a mesoscale WRF model.

[0090] Firstly, the conventional climate input data and terrain data need to be determined; secondly, the longitude and latitude of the reference point of the wind field to be simulated and the simulation time period need to be determined; then, the calculation domain range, grid scale, nesting mode, parameterization scheme configuration, monitoring point position and monitoring variables of the mesoscale WRF simulation need to be determined; finally, the mesoscale numerical simulation of the complex terrain wind field is completed based on the above configuration, and the result is analyzed.

[0091] Step 2, obtain external field observation data and high-precision terrain data, combine high-precision terrain and high-resolution grid, and further improve the accuracy of mesoscale numerical simulation of complex terrain wind field based on four-dimensional variational assimilation method;

[0092] First, obtain external field observation data; second, use observation data to verify the mesoscale numerical simulation results in step 1, if the accuracy meets the requirements, skip step 2; otherwise, combine high-precision terrain and high-resolution grid, and further improve the accuracy of mesoscale numerical simulation of complex terrain wind field based on four-dimensional variational assimilation method; finally, after data assimilation, use observation data to verify the assimilation effect.

[0093] Step 3, establish a meso-micro scale coupling simulation framework based on WRF-CFD model coupling, and perform high-fidelity CFD numerical simulation on the target wind field;

[0094] First, establish a meso-micro scale coupling simulation framework based on WRF-CFD model coupling; then, perform high-fidelity CFD numerical simulation on the target wind field; finally, verify the WRF-CFD numerical simulation results based on external field measurement data.

[0095] Step 4, apply four-dimensional variational assimilation technology to the CFD simulation process to further improve the accuracy of WRF-CFD coupling simulation. If the accuracy of WRF-CFD results in step 3 meets the requirements, skip step 4; otherwise, apply four-dimensional variational assimilation technology to the CFD simulation process to further improve the accuracy of WRF-CFD coupling simulation.

[0096] The preferred examples of each step are described in detail below.

[0097] I. Obtain routine meteorological and terrain data, determine the reference point, and perform numerical simulation of complex terrain wind field based on mesoscale WRF model.

[0098] First, the routine climate input data and terrain data need to be determined. Meteorological data can be selected from the fifth generation data (ERA5) of the European Center for Medium-Range Weather Forecasts (ECMWF), with a spatial resolution of 0.25° and a time resolution of 1 hour; static data of routine terrain categories and terrain height can be selected from data of the United States Geological Survey (USGS), with a maximum spatial resolution of 30".

[0099] Second, determine the latitude and longitude of the reference point of the wind field to be simulated and the simulation time period. The reference point position is selected as the center of the wind field; the simulation time period needs to remove the spin-up time of the mesoscale WRF model, which is usually recommended to be 6-12 hours, and for more complex cases, the time can be extended to 24 hours.

[0100] Then, the calculation domain range, grid scale, nesting method, parameterization scheme configuration, monitoring point location, and monitoring variables of the mesoscale WRF simulation are determined. The calculation domain range is determined according to the reference point location and the size of the target wind field; the grid scale of the innermost nested domain can be 1 km or 3 km, the grid scale of the outermost nested domain can be 20-30 km, and three or four layers of nesting can be used; the two-way nesting method is selected for the nesting method; the parameterization scheme includes microphysical processes, shortwave and longwave radiation, surface and land surface processes, urban and ocean models, planetary boundary layer schemes, and cumulus parameterization processes. For mesoscale numerical simulation of complex terrain wind field, for the microphysical process scheme, the Thompson format (Thompson graupel scheme) and WDM6 are preferred; for shortwave and longwave radiation, the RRTMG+RRTMG scheme combination is preferred; for the selection of the near-surface layer scheme, the Monin-Obukhov scheme is generally used, and the MYJ planetary boundary layer scheme is used in combination; for China's local area, the influence of humidity is larger, at this time, the MYNN 2.5 level TKE planetary layer scheme can be selected, and the MYNN near-surface layer scheme is used in combination; for the land surface process and cumulus convection parameterization scheme, the Noah format and Kain-Fritsch (new Eta) are generally recommended. The monitoring location is the wind tower location, and the monitoring variables can be wind speed, pressure, and temperature. Finally, based on the above configuration, the mesoscale numerical simulation of complex terrain wind field is completed, and the results are analyzed.

[0101] II. Obtain external observation data and high-precision terrain data, combine high-precision terrain and high-resolution grid, and further improve the accuracy of mesoscale numerical simulation of complex terrain wind field based on four-dimensional variational assimilation method.

[0102] In step 2, the external observation data should include the latitude and longitude of the observation site and the altitude of the observation location (or the height above ground of each observation layer), and the observation data includes wind speed, wind direction, temperature, humidity, and pressure variables; the integrity and reliability of the observation data should be ensured to ensure that the time resolution of the observation data is less than or equal to 10 min.

[0103] The observation data is used to verify the mesoscale numerical simulation results in step 1, if the accuracy meets the requirements, then step 2 is skipped; otherwise, high-precision terrain and high-resolution grid are combined, and the accuracy of mesoscale numerical simulation of complex terrain wind field is further improved based on four-dimensional variational assimilation method.

[0104] As shown in Figure 1a , the Observational Nudging method is used in the mesoscale WRF model for data assimilation, and its core expression is:

[0105] ;

[0106] where i is the index of the observation data, x, y, z are the spatial dimension coordinates, t represents the simulation time, q represents the assimilated physical quantity (could be velocity, temperature, humidity, etc.), μ is the dry air static pressure, q0 is the observed value of the physical quantity, q m (x i ,y i ,z i ,t) is the predicted value of the WRF model, G q is the assimilation strength of q, N is the total number of observation data, W q is the weighting function based on the distance between the grid cell and the observation point. W t , W z , and W xy describe the assimilation strength in time and space. The assimilation spatial weight function W xy and W z define the spatial assimilation weight, while the time weight function W t defines the duration and the weight strength in time. Generally, W t increases gradually from 0 to 1 and then decreases gradually to 0.

[0107] The assimilation horizontal weight function is defined as follows:

[0108] ;

[0109] where R is the maximum horizontal influence radius, and D is the horizontal distance between the grid point and the observation point, as shown in Figure 1b .

[0110] The time weight function is defined as follows, as shown in Figure 1c

[0111] ;

[0112] where t0 is the observation time point, t is the simulation time, is the time window length of the observation data (in hours).

[0113] Allows the specification of how a variable is relaxed for assimilation on the vertical profile with full weight and / or a decreasing function relative to the surface or the top of the boundary layer and can be set for different conditions (stable or unstable). The vertical weight is usually set very small so that the data is only assimilated on its own eta level.

[0114] ​When the four-dimensional variational assimilation technique is used in the mesoscale WRF model, the observation data format needs to be first converted into the little_r format, and then further converted into the OBS_DOMAIN*01 format (where * is the number of the nested domain to be assimilated) through the OBSGRID code. Then, the data assimilation control parameters are configured, including the start and end times of the assimilation, the specific nested domain to be assimilated, the variables to be assimilated, the horizontal and vertical influence ranges during data assimilation, the data assimilation time window, and the data assimilation intensity. Among them, the horizontal influence range can be 180-240 km; in the multi-layer assimilation process, the vertical direction influence can be controlled by the observation data pressure interpolation, and the interpolation will not be performed on the model levels whose observation vertical interval is greater than or equal to the maximum pressure interval (obs_max_sndng_gap); the data assimilation time window length can be 0.11 h (corresponding to a 10 min time resolution of the observation data); and the data assimilation intensity can be 6.0×10 -4 .

[0115] During data assimilation, the high-precision terrain data SRTM 30 of the Space Shuttle Radar Topography Mission is used to replace the USGS data in the innermost nested domain, and the grid resolution of the innermost nested domain can be 200-333 m. Finally, after the data assimilation is completed, the observation data is used to verify the assimilation effect.

[0116] III. Establishing a meso-micro scale coupling simulation framework based on WRF-CFD model coupling to perform high-fidelity CFD numerical simulation on the target wind field.

[0117] In step 3, first, a meso-micro scale coupling simulation framework based on WRF-CFD model coupling is established, Figure 2 is the establishment process of the coupling boundary under the WRF-CFD model coupling framework. The boundary coupling method is used for WRF-CFD data transmission. The spatial and temporal sequences of wind speed, temperature, and turbulent kinetic energy and other parameters corresponding to the boundary of the microscale calculation domain are extracted from the WRF simulation results, and then they are imported into the microscale calculation domain boundary through time and space interpolation for microscale calculation. The establishment of the coupling boundary is completed in the following steps:

[0118] 1) Determine the latitude and longitude and coordinates of the reference point, which is generally located at the center point of the mesoscale calculation domain and the microscale calculation domain;

[0119] 2) Determine the boundary of the microscale calculation domain, which is generally determined according to the height of the mountain and the thickness of the atmospheric boundary layer;

[0120] 3) Extract the boundary coordinates of the microscale calculation domain, which is completed by using the OpenFOAM data monitoring tool combined with Python scripts;

[0121] 4) Extract the wind speed, turbulent kinetic energy, temperature, pressure and other information corresponding to the boundary coordinates from the mesoscale numerical simulation results. Here, the (x, y, z) coordinates need to be converted to longitude, latitude, and altitude. Both coordinates are in the UTM projection system. Due to the large difference between mesoscale and microscale grids, spatial interpolation is required. This is done using Fortran scripts combined with Python scripts.

[0122] 5) In the microscale calculation, the boundary variables derived in step 4) are added to the microscale calculation domain boundary to complete the CFD numerical simulation. Due to the difference in time step between mesoscale and microscale simulations, time interpolation is required.

[0123] For any two time steps, linear interpolation is used to assign values. Mesoscale data extraction is based on altitude. Spatial interpolation uses triangular mesh / linear interpolation.

[0124] To ensure consistency between mesoscale simulation output data and microscale simulation boundary input data, meso-microscale terrain fusion is required.

[0125] To address the issues of insufficient mesoscale numerical simulation fluctuations and slow development of near-surface turbulence, artificial synthetic turbulence based on measured data is added to the import and export boundaries.

[0126] The main operation steps for microscale calculation domain grid generation are as follows:

[0127] 1) Prepare the terrain file (convert the.map format terrain file to the.stl format terrain file in advance);

[0128] 2) Prepare the blockMeshDict and snappyHexMeshDict files;

[0129] 3) Run the script to generate the grid file (after grid generation, perform grid quality check).

[0130] The east, south, west, and north of the microscale calculation domain are import and export boundary conditions, and the vertical top and bottom are top and bottom boundaries. In the import and export boundary conditions, the variables U and T use time-varying boundary conditions.

[0131] The time is discretized by backward difference method (second order, implicit), the gradient term is discretized by Gauss linear interpolation (second order, unbounded), the convection and divergence terms are discretized by Gauss linear interpolation (second order, unbounded), and the Laplacian term is discretized by Gauss linear corrected interpolation (explicit, non-orthogonal grid correction).

[0132] The linear equation system is solved by iterative methods based on conjugate gradient (PCG for variable p_rgh) and bi-conjugate gradient (PBiCG for variables U, T, k, nuTilda). The preconditioner is geometric agglomeration algebraic multigrid (GAMG for variable p_rgh) and diagonal incomplete LU (DILU for variables U, T, k, nuTilda). In addition, the smoothing solver (DICGaussSeidel for variable p_rgh) is used.

[0133] The PISO-SIMPLE is used to solve the momentum and pressure equations implicitly, the Rhie-Chow interpolation method is used in the predictor-corrector method, and the temperature equation is added in the correction cycle. Then, the target wind field is simulated by high-fidelity CFD numerical simulation. Finally, the WRF-CFD numerical simulation results are verified based on the measured data of the external field.

[0134] Four, apply four-dimensional variational assimilation technology to the CFD simulation process to further improve the accuracy of WRF-CFD coupled simulation.

[0135] If the accuracy of WRF-CFD results in step 3 meets the requirements, step 4 is skipped; otherwise, apply four-dimensional variational assimilation technology to the CFD simulation process to further improve the accuracy of WRF-CFD coupled simulation.

[0136] In step 4, when the four-dimensional variational assimilation technology is applied to the CFD simulation, a source term F is added to the momentum equation to correct the velocity field U. The calculation of the source term is based on the time, space weight, and the difference between the current velocity field and the target velocity field. At this time, the momentum equation expression is:

[0137] ;

[0138] In the formula: is the velocity component, subscript respectively represent the flow direction, transverse direction and vertical direction, for the coordinate components, for time, for fluid density, for static pressure, represents potential temperature, characterized by horizontal averaging. Other parameters: for gravitational acceleration, is the reference potential temperature, represents the Coriolis force coefficient, for the average angular velocity of the Earth's rotation, for the local latitude, , represent the streamwise and crosswise components of the geostrophic wind, respectively, is the Dirac function, j = 1, 2, 3, when , , and 0 otherwise. for stress, is the source term for velocity correction.

[0139] The terms in the momentum equation represent:

[0140] the unsteady term, the convection term, the pressure gradient term, the stress (viscosity + SGS / Reynolds), the buoyancy, the geostrophic deflection force, the source term.

[0141] During the CFD simulation process, the flow field velocity is adjusted to the target wind speed profile by updating the source term F. See Figure 3 , the application process of four-dimensional variational assimilation technology (Observational Nudging) in micro-scale CFD calculation includes the following steps:

[0142] 1) Calculation of target position average wind speed:

[0143] Calculate the weighted average wind speed within the specified grid volume at the target position:

[0144] ;

[0145] In the formula, represents the wind speed vector in the i-th grid element corresponding to the j-th assimilation point, represents the volume in the i-th grid element corresponding to the j-th assimilation point, j is a certain point that needs to be assimilated, and m is the number of grids within the specified volume.

[0146] Remove vertical component (Assume adjustment only acts on horizontal direction):

[0147] ;

[0148] where, is the unit normal vector in the vertical direction.

[0149] 2) Target position real-time wind speed reading:

[0150] Read the wind speed vector of the target position at the latest time point from the wind speed time series .

[0151] 3) Source term adjustment amount calculation:

[0152] Based on the difference between the target wind speed and the current average wind speed, calculate the wind speed adjustment amount ds, and normalize it through rAU (the inverse of the diagonal matrix of the momentum equation):

[0153] ;

[0154] ;

[0155] ;

[0156] where, α is the relaxation factor, is the volume-weighted average value of rAU.

[0157] 4) Spatial and temporal weight function calculation:

[0158] The weight w j is composed of three parts:

[0159] A. Horizontal weight (inverse proportional to horizontal distance, consistent with the Observational Nudging method):

[0160] ;

[0161] where, r xy is the maximum horizontal influence radius, d xy is the horizontal distance between the grid point and the observation point.

[0162] B. Vertical weight (linear decay, inconsistent with the Observational Nudging method):

[0163] ;

[0164] where, r z is the maximum vertical influence range, d zis the vertical distance between the grid point and the observation point.

[0165] C. Time weight (piecewise function, consistent with the Observational Nudging method):

[0166] ;

[0167] where t0 is the observation time point, t is the simulation time, is the length of the time window of the observation data (in seconds).

[0168] The total weight is:

[0169] ;

[0170] where, , , respectively represent the assimilation space weight function and the time weight function of the jth assimilation point; the maximum horizontal influence radius r xy The maximum vertical influence range r z The time window is determined to be the same height as the microscale calculation, and the determination is consistent with the Observational Nudging method in the mesoscale WRF model.

[0171] 5) Source term update:

[0172] The adjustment amount is distributed to the grid according to the weight, and the F and velocity field U are updated:

[0173] ;

[0174] ;

[0175] The flux field φ is updated synchronously:

[0176] ;

[0177] where n is the number of grids in the calculation domain, represents the face interpolation field of the inverse of the diagonal matrix of the momentum equation rAU, which is used to construct the pressure correction equation to ensure the numerical stability of the velocity-pressure coupling, fvc represents the finite volume convection, interpolate represents the interpolation function, which is used to interpolate the physical quantity at the center of the grid cell to the grid interface, represents the area vector of the grid interface.

[0178] Examples

[0179] Select a complex terrain wind farm (such as Figure 4As the research object, this wind farm is a typical case for studying the characteristics of wind flow in complex terrain. The terrain in the area where the wind farm is located is a series of hills, with the highest elevation of 1824 meters and the lowest elevation of 1222 meters. The main vegetation types in the wind farm include farmland, grassland, shrub land, and forest, etc. The average roughness length of this area is 0.0624 meters.

[0180] Step 1, Obtain regular meteorological and terrain data, determine the reference point, and conduct numerical simulation of complex terrain wind field based on the mesoscale WRF model

[0181] Firstly, the regular climate input data and terrain data need to be determined. The meteorological data is selected from the fifth generation data (ERA5) of the European Centre for Medium-Range Weather Forecasts (ECMWF), with a spatial resolution of 0.25° and a temporal resolution of 1 hour; the static data of regular terrain categories and terrain height can be selected from the data of the United States Geological Survey (USGS) (as shown in Figure 5a ), with the highest spatial resolution of 30".

[0182] Secondly, the latitude and longitude of the reference point of the wind field to be simulated and the simulation time period are determined. The coordinates of the reference point location are (37.2286°N, 112.7075°E); the time period of WRF simulation is from 08:00 on December 20, 2022 to 00:00 on December 23, 2022 (local time, UTC+8), and the WRF simulation time resolution is 60 s, 20 s, 6.67 s and 2.22 s respectively. For this case, the spin-up time is selected as 16 hours.

[0183] Then, the calculation domain range, grid scale, nesting method, parameterization scheme configuration, monitoring point location and monitoring variables of mesoscale WRF simulation are determined. For this case, the grid scale of the innermost nesting domain is 1 km, and the grid scale of the outermost nesting domain is 27 km; the projection method is set to lambert; the spatial horizontal resolution is 27 km, 9 km, 3 km and 1 km respectively, and 4 layers of nesting are used; the nesting method is selected as bidirectional nesting method; for this case, the microphysical process scheme is selected as WDM6; for shortwave and longwave radiation, the RRTMG+ RRTMG scheme combination is used; for the selection of near-surface layer scheme, the Monin-Obukhov scheme is used, and the MYJ planetary boundary layer scheme is used in combination; for the land surface process and cumulus convection parameterization scheme, the Noah format and Kain-Fritsch (new Eta) are used. The monitoring location is the wind tower location (the radar location), and the monitoring variables are selected as wind speed, pressure, potential temperature, etc.

[0184] Finally, based on the above configuration, the mesoscale numerical simulation of complex terrain wind field is completed, and the results are analyzed.

[0185] Step 2, Obtain external field observation data and high-precision terrain data, combine high-precision terrain and high-resolution grid, and further improve the accuracy of mesoscale numerical simulation of complex terrain wind field based on four-dimensional variational assimilation method

[0186] First, obtain external field observation data. The observation data used this time comes from a wind laser radar located at (112.70750004°E, 37.22866700°N) with a terrain height of 1670 meters, and its sampling frequency is 1Hz. The observation data includes temperature, pressure, humidity, wind speed and wind direction. The measurement heights of the laser radar are 40 meters, 44 meters, 47 meters, 58 meters, 61 meters, 73 meters, 91 meters, 101 meters, 113 meters and 121 meters. The time range of the data used this time is from 0:00 on December 21, 2022 to 0:00 on December 23, 2022 (local time, UTC+8).

[0187] Second, use observation data to verify the mesoscale numerical simulation results in step 1, and find that the accuracy is low, combined with high-precision terrain (such as Figure 5b ) and high-resolution grid (the outermost nested domain grid scale is 25km, the innermost nested domain grid scale is 200m, the vertical height is 56 layers, and the near-surface area is densified, with 10 layers below 100 meters, 10 layers in the range of 100-500 meters, 5 layers in the range of 500-1000 meters, and 31 layers above 1000 meters. Other configurations are the same as in step 1), based on four-dimensional variational assimilation method to further improve the accuracy of mesoscale numerical simulation of complex terrain wind field.

[0188] The assimilation start time is 2022.12.21.17:00 (UTC time) and the assimilation end time is 2022.12.23.15:00 (UTC time). Data assimilation is performed in four layers of nested domains. The maximum horizontal influence radius is 180km in the third and fourth layers of nested domains, and 240km in the first and second layers of nested domains. The data assimilation time window length can be 0.11h (corresponding to the observation data with a time resolution of 10min); the data assimilation strength is 6.0×10 -4 .

[0189] During data assimilation, the innermost nested domain uses the high-precision terrain data SRTM 30 of the Space Shuttle Radar Topography Mission to replace the USGS data, and the innermost grid resolution uses 200m, as shown in Figure 5b .

[0190] Finally, after data assimilation, the observation data is used to verify the assimilation effect.

[0191] Step 3, establish a meso-micro scale coupling simulation framework based on WRF-CFD model coupling, and perform high-fidelity CFD numerical simulation on the target wind field

[0192] First, a meso-micro scale coupling simulation framework based on WRF-CFD model coupling is established.

[0193] The boundary coupling method is used for WRF-CFD data transmission. The spatial and temporal sequence of wind speed, temperature and turbulent kinetic energy and other parameters corresponding to the boundary of the micro scale calculation domain is extracted from the WRF simulation results, and then it is imported into the micro scale calculation domain boundary through time and space interpolation for micro scale calculation.

[0194] For any two time steps, linear interpolation method is adopted for assignment. The mesoscale data extraction is based on altitude. The spatial interpolation adopts triangular net / linear interpolation method.

[0195] In order to ensure the consistency of mesoscale simulation output data and microscale simulation boundary input data, meso-micro scale terrain fusion is needed.

[0196] In order to solve the problems of insufficient mesoscale numerical simulation fluctuation and slow development of near-surface turbulent flow, artificial synthetic turbulence based on measured data is added on the basis of import and export boundary.

[0197] The main operation steps of micro scale calculation domain grid (such as Figure 6a and 6b ) generation technology are as follows:

[0198] 1) Prepare the terrain file (need to convert the.map format terrain file to.stl format terrain file in advance);

[0199] 2) Prepare blockMeshDict and snappyHexMeshDict files;

[0200] 3) Run the script to generate the grid file (after grid generation, perform grid quality check).

[0201] For this case, the micro scale calculation domain height value is 2 km, and the horizontal direction size is set to 5 km x 5 km. The uniform background grid horizontal resolution is 16.5 m, and the vertical resolution is 10 m. The grid is densified in the near-surface range covering the mountain, and the densified area grid horizontal resolution is 4 m, and the vertical resolution is 2.5 m.

[0202] The east, south, west and north of the micro scale calculation domain are import and export boundary conditions, and the top and bottom are top and bottom boundaries respectively. In the import and export boundary conditions, the variables U and T adopt time-varying boundary conditions.

[0203] The time is discretized using the backward difference method (second order, implicit), the gradient term is discretized using the Gauss linear interpolation scheme (second order accuracy, unbounded), the numerical scheme for the convection and divergence terms is basically the same, and the Gauss linear interpolation scheme (Gauss linear) is used, and the limited linear 1 is used in the upwind direction of the linear scheme, such as div(phi,nuTilda), and the Laplace term is discretized using the Gauss linear interpolation correction scheme (Gauss linear corrected, explicit non-orthogonal grid correction).

[0204] The linear equation solving method mainly includes iterative solving based on the conjugate gradient method (PCG, for variable p_rgh) and iterative solving based on the bi-conjugate gradient method (PBiCG, for variables U, T, k, nuTilda, etc.). The preconditioner includes the algebraic multigrid preconditioner (GAMG, for variable p_rgh), the diagonal-based incomplete LU preconditioner (DILU, for variables U, T, k, nuTilda, etc.). In addition, the fairing solver (DICGaussSeidel method for variable p_rgh) is also used.

[0205] The solver uses PISO-SIMPLE to "implicitly" solve the momentum and pressure equations, the Rhie-Chow interpolation method is used in the prediction-correction method, and the temperature equation is added in the correction cycle.

[0206] Then, high-fidelity CFD numerical simulation is performed on the target wind field;

[0207] Finally, the WRF-CFD numerical simulation results are verified based on the measured data of the external field.

[0208] Step 4: Apply four-dimensional variational assimilation technology to the CFD simulation process to further improve the accuracy of WRF-CFD coupled simulation

[0209] For this case, the four-dimensional variational assimilation technology is applied to the CFD simulation process to further improve the accuracy of WRF-CFD coupled simulation.

[0210] When the four-dimensional variational assimilation technology is applied to the CFD simulation, a source term F is added to the momentum equation to correct the velocity field (U). The calculation of the source term is based on the time, space weight, and the difference between the current velocity field and the target velocity field.

[0211] In this case, the observation data (wind speed vector) at radar heights of 40, 101 and 143 meters is used for assimilation. The average velocity within a 20-meter radius sphere at the observation point is solved.

[0212] Maximum horizontal influence radius r xy 3000m in this case. Maximum vertical influence range r z 1500m in this case. Observation time point t0 every 10 min; the length of the time window of observation data In seconds, 1200.5s in this case (for accelerated convergence).

[0213] After analysis Figure 7a , Figure 7b and Figure 7c It can be seen that when there is no observation data assimilation, the simulation results of wind speed and wind direction based on the mesoscale WRF model simulation of complex terrain wind field have large differences with the observation data, so data assimilation is needed to improve the numerical simulation accuracy. When combined with high-precision terrain and high-resolution grid, and using four-dimensional variational assimilation technology, the average absolute error of wind speed, wind direction and turbulence intensity at the radar position is reduced by 40.1%, 27.4% and 6.1% (compared with the mesoscale WRF numerical simulation method without observation data assimilation); At the same time, it is found that there is a phenomenon of insufficient turbulence generation in the mesoscale WRF model. In order to further improve the simulation accuracy of complex terrain wind field, reflect the capture of high-precision terrain on flow details, add the assimilated WRF results into the CFD boundary to establish a complex terrain wind field numerical simulation method based on WRF-CFD coupling framework, at this time the average absolute error of wind speed, wind direction and turbulence intensity at the radar position is reduced by 51.1%, 45.5% and 36.2% (compared with the mesoscale WRF numerical simulation method without observation data assimilation), Especially the simulation accuracy of turbulence intensity is improved obviously. Further, after applying four-dimensional variational assimilation technology to the CFD simulation process, the average absolute error of wind speed, wind direction and turbulence intensity at the radar position is reduced by 97.6%, 90.7% and 86% (compared with the mesoscale WRF numerical simulation method without observation data assimilation), At this time, the average absolute error of wind speed, wind direction and turbulence intensity is 0.15m / s, 3.2°, 0.04, respectively, while the average absolute error of wind speed, wind direction and turbulence intensity in the mesoscale WRF numerical simulation method without observation data assimilation is 6.22m / s, 34.4°, 0.29, respectively.

[0214] In summary, the complex terrain wind field simulation technology of the application can simultaneously integrate data assimilation method into WRF process and microscale CFD simulation, and fully consider factors such as grid resolution, terrain resolution, assimilation frequency and unsteady effect, which can greatly improve the numerical simulation accuracy of complex terrain wind field, effectively improve the accuracy of wind resource assessment, and lay a solid foundation for wind farm planning and operation control.

[0215] Compared with the wind field measured data, it is found that the real-time assimilation method of complex terrain wind field multi-scale numerical simulation fusing external field measurement data can well simulate the wind speed, wind direction and turbulence intensity in complex terrain wind field, and has good consistency with the observation data. The data assimilation method has the advantages of simple form, easy coding, high efficiency and high calculation precision, and has good prediction accuracy for the spatial and temporal distribution of wind speed, wind direction and turbulence, and the calculation accuracy is better than the numerical simulation results based on the coupling method of WRF model and WRF-CFD model.

[0216] Although preferred embodiments of the application have been described, those skilled in the art will be able to make additional modifications and variations to the embodiments without departing from the spirit and scope of the application. Accordingly, the appended claims are intended to encompass all such modifications and variations as falling within the scope of the application.

[0217] Obviously, various modifications and changes can be made to the present application by those skilled in the art without departing from the spirit and scope of the present application. Thus, it is intended that the present application embrace all such modifications and changes as fall within the scope of the claims and their equivalents.

Claims

1. A real-time assimilation method of complex terrain wind field multi-scale numerical simulation fusing external field measurement data, characterized in that, The method comprises the following steps: S1, acquiring conventional meteorological and terrain data, determining a reference point, and simulating a complex terrain wind field based on a mesoscale WRF model; judging the accuracy of the mesoscale numerical simulation of the complex terrain wind field, and if the first preset accuracy requirement is met, proceeding to step S3, otherwise, proceeding to step S2; S2, acquiring external field observation data and high-precision terrain data, combining high-precision terrain and high-resolution grids, and improving the accuracy of the mesoscale numerical simulation of the complex terrain wind field to meet the first preset accuracy requirement based on a four-dimensional variational assimilation method; S3, establishing a meso-micro scale coupling simulation framework based on WRF-CFD model coupling, and performing high-fidelity CFD numerical simulation on the target wind field; judging the accuracy of the CFD numerical simulation, and if the second preset accuracy requirement is met, ending the process, otherwise, proceeding to step S4; S4, applying four-dimensional variational assimilation technology to the CFD simulation process to improve the accuracy of the CFD numerical simulation to meet the second preset accuracy requirement; In step S2, the accuracy of the mesoscale numerical simulation of the complex terrain wind field is improved based on the four-dimensional variational assimilation method, and the data assimilation expression is: where i is the index of observation data, x, y, z are spatial dimension coordinates, t represents simulation time, q represents assimilated physical quantity, μ is the static pressure of dry air, q0 is the observed value of physical quantity, q m (x i ,y i ,z i ,t) is the prediction value of WRF model, G q is the assimilation intensity of q, N is the total number of observation data, W q is the weighting function based on the distance between grid cells and observation points; W t , W z and W xy describe the assimilation intensity of time and space; W xy and W z are the assimilation spatial weight functions, W t is the time weight function; In the formula, R is the maximum horizontal influence radius, D is the horizontal distance between the grid point and the observation point, t0 is the observation time point, t is the model time, τ is the time window length of the observation data; W t The value range of W is 0 to 1. Step S3 further comprises: S31, establishing a meso-micro scale coupling simulation framework based on WRF-CFD model coupling, and performing WRF-CFD data transfer by boundary coupling; specifically, extracting the wind speed, temperature and turbulent kinetic energy parameters corresponding to the boundary of the microscale calculation domain from the WRF simulation results, and importing the extracted data into the boundary of the CFD simulation calculation domain through time and space interpolation for microscale calculation; S32, performing high-fidelity CFD numerical simulation on the target wind field; S33, verifying the WRF-CFD numerical simulation results based on external field measured data.

2. The complex terrain wind field multiscale numerical simulation real-time assimilation method of fusing external field measurement data according to claim 1, characterized in that, Step S1 further comprises the following steps: S11, determining conventional climate input data and terrain data; S12, selecting the wind field center as the reference point of the wind field to be simulated, and acquiring its latitude and longitude; removing the start time of the mesoscale WRF model, and determining the simulation time period of the wind field to be simulated; S13, determining the calculation domain range, grid scale, nesting method, parameterization scheme configuration, monitoring point position and monitoring variable of the mesoscale WRF simulation; S14, performing mesoscale numerical simulation of the complex terrain wind field, and performing result analysis.

3. The complex terrain wind field multiscale numerical simulation real-time assimilation method of fusing external field measurement data according to claim 2, characterized in that, In step S13, the calculation domain range is determined according to the reference point position and the size of the target wind field; the grid scale and time scale of each nested domain are determined according to the complexity of the underlying surface terrain; The nesting method is selected as bidirectional nesting; the parameterization scheme includes microphysical process, shortwave and longwave radiation, surface and land surface process, city and ocean model, planetary boundary layer scheme and cumulus parameterization process; The monitoring position is the position of the wind measurement tower, and the monitoring variables include wind speed, pressure and potential temperature.

4. The complex terrain wind field multiscale numerical simulation real-time assimilation method of fusing external field measurement data according to claim 1, characterized in that, In step S1, the process of judging the accuracy of the mesoscale numerical simulation of the complex terrain wind field comprises: Acquiring external field observation data, including the latitude and longitude and altitude of the observation site, and the wind speed, wind direction, temperature, humidity and air pressure variables of each observation site; and verifying whether the accuracy of the mesoscale numerical simulation of the complex terrain wind field meets the first preset accuracy requirement by using the observation data.

5. The complex terrain wind field multiscale numerical simulation real-time assimilation method of fusing external field measurement data according to claim 1, characterized in that, In step S31, the process of establishing the coupling boundary includes the following steps: S311, determining the longitude and latitude and coordinates of the reference point, which is located at the center point of the mesoscale calculation domain and the microscale calculation domain; S312, determining the boundary of the microscale calculation domain according to the height of the mountain and the thickness of the atmospheric boundary layer; S313, extracting the boundary coordinates of the microscale calculation domain; S314, extracting the wind speed, turbulent kinetic energy, temperature, and pressure information corresponding to the boundary coordinates from the mesoscale numerical simulation results, and performing spatial interpolation and time interpolation on the extracted boundary variables, so that the interpolated boundary variables meet the spatial requirements of the microscale grid and the time requirements of the microscale simulation time step; S315, adding the derived boundary variables to the boundary of the microscale calculation domain to complete the CFD numerical simulation.

6. The complex terrain wind field multiscale numerical simulation real-time assimilation method of fusing external field measurement data according to claim 1, characterized in that, Step S4 further includes: S41, adding a source term F and a modified velocity field U to the momentum equation, and modifying the expression of the momentum equation as: In the formula: u i For velocity components, subscripts i = 1, 2, 3 represent the flow direction, lateral direction, and vertical direction, respectively, and x i Let be the coordinate components, be time, ρ be the fluid density, p be the static pressure, and θ be the potential temperature. The value is represented by the mean value on the horizontal plane; g is the acceleration due to gravity, and θ0 is the reference potential temperature. The Coriolis force coefficient is represented by Ω = 7.292 × 10⁻⁶. -5 rad / s is the average angular velocity of the Earth's rotation. For the local latitude, u G v G These represent the directional and lateral components of the geostrophic wind, respectively; δ ij It is a Dirac function, j = 1, 2, 3, when i = j, δ ij =1, and 0 in other cases; τ ij For stress, F is the source term used for velocity correction; S42, adjusting the flow field velocity to the target wind speed profile by updating the source term F.

7. The complex terrain wind field multiscale numerical simulation real-time assimilation method of fusing external field measurement data according to claim 6, characterized in that, Step S42 further includes: S421, calculating the weighted average wind speed within the specified grid volume at the target position: wherein, represents the wind speed vector in the i-th grid cell corresponding to the j-th assimilation point, represents the volume in the i-th grid cell corresponding to the j-th assimilation point, j is a certain point that needs to be assimilated, and m is the number of grids in the specified volume; after removing the vertical component, we get: In the formula, n up represents a unit normal vector in the vertical direction; S422, read the wind speed vector of the target location at the most recent time point from the wind speed time series S423, calculating the wind speed adjustment amount ds based on the difference between the target wind speed and the current average wind speed, and normalizing it: ds j = ds j - (ds j · n up ) n up ; ds j = a - ds j ; wherein a is a relaxation factor, <rau> vol is the volume-weighted average of the inverse of the diagonal matrix of the momentum equation rAU;< / rau> S424, calculating the spatial and temporal weight functions; wherein the horizontal weight is: where r xy is the maximum horizontal influence radius, d xy is the horizontal distance between the grid point and the observation point; The vertical weight is: where r z is the maximum vertical impact range, d z is the vertical distance between the grid point and the observation point; The time weight is: In the formula, t0 is the observation time point, t is the simulation time, and τ is the time window length of the observation data; The total weight is: wherein respectively denote the assimilation space weight function and the time weight function for the jthassimilation point. S425, distributing the adjustment amount to the grid according to the weight, and updating F and the velocity field U as: Synchronously updating the flux field φ as: where n is the number of grid points in the computational domain, rAU f represents the face interpolation field of the inverse of the diagonal matrix of the momentum equation, rAU, used to construct the pressure correction equation to ensure numerical stability of the velocity-pressure coupling, fvc represents the finite volume convection, interpolate represents the interpolation function used to interpolate the physical quantity at the center of the grid cell to the grid interface, S f represents the area vector of the grid interface.

Images