Method and system for predicting glaze ice taking into account underlying surface and cloud physical parameterization

Abstract

The application discloses a kind of fog rime ice forecast method and system considering underlying surface and cloud physical parameterization, belong to electric power meteorological field, including, collection high space-time resolution land cover data MODIS data, MODIS data splicing and pretreatment;Through the full coupling of MAKKONEN fog rime ice model and the cloud microphysics scheme in WRF-ARW, the small-scale numerical model WRF-Icing-Fog capable of directly calculating fog rime icing is obtained;Through global atmospheric reanalysis data or global forecast system forecast data, the preprocessing system WPS of small-scale model WRF-ARW is obtained;Using WRF-Icing-Fog constructs high-resolution fog rime ice numerical prediction model to obtain the final fog rime ice diameter and fog rime ice mass and other ice results.The cloud microphysics parameterization scheme in WRF-ARW model-Thompson-aero-MP scheme and MAKKONEN fog rime ice model are fully coupled to construct WRF-Icing-Fog fully coupled fog rime ice numerical prediction model, which significantly reduces the calculation steps of offline calculation ice, and directly obtains fog rime icing characteristic data.

Description

Technical Field

[0001] This invention belongs to the field of power meteorology technology, specifically relating to the field of numerical forecasting technology for rime ice accumulation, and more specifically, to a power meteorology rime ice accumulation forecasting method and system that considers the coupling of complex underlying surface and cloud physical parameterization. Background Technology

[0002] The State Grid Corporation of China and meteorological departments primarily focus on the following icing (ice accumulation) characteristics during the transition between cold and warm seasons: average ice diameter (from rime, hoarfrost, or snow), spatial and temporal distribution of icing, and duration of icing. Currently, there are generally three methods for obtaining icing-related data in domestic and international scientific research on icing monitoring and forecasting, as well as in power meteorological operational services:

[0003] I. Obtaining icing observation data from observation points through meteorological departments or the State Grid Corporation of China. This is the most accurate method to obtain data on the diameter or mass of icing at a specific point. However, the spatial representativeness of station observations is insufficient, making large-scale icing monitoring impossible. Furthermore, this method requires significant investment of manpower and resources, and it is difficult to obtain long-term observation time series. In complex regions of China, especially those with complex underlying surfaces, conducting icing observations in uninhabited areas is extremely challenging. Overcoming the problem of spatiotemporal continuity of icing observation data is crucial, and establishing and maintaining these stations requires substantial financial and human resources.

[0004] Second, rime ice models can be calculated using gridded elements such as surface precipitation, vertical multi-level temperature, vertical multi-level liquid water content, wind direction, and wind speed from publicly available global atmospheric reanalysis data products from various World Meteorological Centres under the WMO (World Meteorological Organization) and meteorological departments at home and abroad (such as in Europe and the United States or China). Examples include, but are not limited to, the MAKKONEN model. More widely used datasets include the latest generation of global atmospheric reanalysis data ERA5-Reanalysis from the ECMWF (European Centre for Medium-Range Weather Forecasts), and the FNL (Final) dataset from the NCEP (National Centers for Environmental Prediction) or the US National Oceanic and Atmospheric Administration. These datasets provide spatiotemporally continuous gridded elements such as surface precipitation, vertical multi-level temperature, vertical multi-level liquid water content, wind direction, and wind speed over historical long-term series (longer than 20 years) for offline driving of rime ice models, obtaining relatively long-term rime ice datasets. However, global data has the following obvious drawbacks: global atmospheric reanalysis data is spatially coarse, generally ranging from 25km to 100km, which cannot resolve the heterogeneity of meteorological elements caused by complex underlying surfaces, and thus cannot meet the needs of refined monitoring of icing disasters in national power meteorology.

[0005] III. Utilizing data from small- and medium-scale numerical weather prediction models to drive rime ice models for post-icing forecasting or prediction of icing diameter. Currently, both domestically and internationally, offline coupling methods are used for icing monitoring and forecasting. A common approach is to use the small- and medium-scale forecasting model WRF-ARW (Advanced Research Weather Research and Forecasting) to forecast regional meteorological conditions, obtaining high-resolution data on meteorological elements such as precipitation, liquid water content, temperature, wind direction, and wind speed. This data is then used to drive rime ice models offline to obtain rime ice detection or forecasting data for relatively large areas. The MAKKONEN rime ice model is frequently used. The calculation process requires first running WRF-ARW, then extracting the required meteorological elements from the WRF-ARW output, and finally driving the rime ice model to obtain icing products. This process is multifaceted and complex.

[0006] Existing technical document 1 (CN112711919A) discloses a method and system for forecasting icing on a conductor based on coupling with small- and medium-scale models. However, the forecasting method described in existing technical document 1 uses meteorological elements calculated and output by small- and medium-scale models as data input to the conductor icing model. This is a traditional one-way coupling calculation method. The meteorological elements calculated and output by small- and medium-scale models are then input into the icing model, which cannot achieve direct output of icing models at different altitude levels. The process is cumbersome and involves many steps.

[0007] Prior art document 2 (CN105184407B) discloses a method for predicting the icing growth of transmission lines based on an atmospheric numerical model. However, the prediction method described in prior art document 2 can only output the icing growth driven by near-surface meteorological elements, and fails to output the icing situation at other vertical heights. In addition, this invention also uses meteorological elements calculated and output by small- and medium-scale models as data input for the conductor icing model, which is also a traditional one-way coupling calculation method with many redundant and complicated steps. Summary of the Invention

[0008] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0009] In view of the above-mentioned existing problems, the purpose of this invention is to provide a numerical prediction method and system for rime ice accumulation that takes into account complex underlying surfaces. This method can solve the technical problems of limited observation data and spatiotemporal discontinuity in traditional rime ice accumulation data products, as well as the low resolution of rime ice accumulation products calculated by rime ice accumulation models driven by publicly available global atmospheric reanalysis products.

[0010] To solve the above-mentioned technical problems, the present invention provides the following technical solution.

[0011] A first aspect of the present invention provides a method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters, comprising the following steps:

[0012] High-resolution MODIS land cover data were collected, stitched together, and preprocessed to obtain a high-precision underlying surface dataset.

[0013] By fully coupling the MAKKONEN hoarfrost and ice accumulation model with the cloud microphysical parameterization scheme in the WRF-ARW model, a small-to-medium scale numerical model WRF-Icing-Fog, which can directly perform hoarfrost and ice accumulation calculations, is obtained.

[0014] Collect atmospheric reanalysis meteorological data or forecasting system data in accordance with the data requirements of the small- and medium-scale numerical model WRF-Icing-Fog.

[0015] The original static underlying surface data of the WRF-ARW model is replaced with the high-precision underlying surface dataset. Then, the atmospheric reanalysis meteorological data or forecast system data are processed by the WRF-ARW preprocessing system WPS to compile and generate a regional high-resolution numerical forecast model for rime ice and icing. This results in regional, high spatiotemporal resolution simulation or prediction results for rime ice and icing that take into account complex underlying surfaces.

[0016] Preferably, the preprocessing includes: projection transformation, resampling, and binary format conversion.

[0017] The projection transformation is expressed by the following formula (1):

[0018] C(x, y) = LU(i, j) (1)

[0019] In the formula:

[0020] C represents the original land cover data, (x, y) represents the location coordinates of the raster data after projection transformation, LU represents the projection function, and (i, j) represents the location coordinates of the raster data before projection transformation.

[0021] The original land cover data C is resampled to a new raster using the four-point interpolation function IN, as shown in the following formula (2):

[0022] C′(m,n)=IN[C(x,y)] (2)

[0023] In the formula:

[0024] C(x, y) represents the original land cover data at location (x, y), IN represents the four-point interpolation function, and C′(m, n) represents the resampled land cover raster data;

[0025] The m×n raster matrix is ​​converted into a WRF-ARW pattern recognition binary file A using the transformation function Tr, as shown in the following formula (3):

[0026] A = Tr(C′) (3)

[0027] In the formula:

[0028] A represents the binary file that WRF-ARW mode can recognize, and Tr represents the conversion function used to convert it into a binary file that WRF-ARW mode can recognize.

[0029] Preferably, the MAKKONEN hoarfrost icing model is fully coupled with the cloud microphysical parameterization scheme in the WRF-ARW model to obtain the small-to-medium scale numerical model WRF-Icing-Fog, which can directly perform hoarfrost icing calculations.

[0030] Meteorological element data for each layer of the vertical layer of the computational model based on the Thompson-aero-MP cloud microphysics scheme in WRF-ARW;

[0031] The MAKKONEN rime ice accumulation model is coupled to calculate ice thickness. Meteorological data for each layer are used as input to the MAKKONEN rime ice accumulation model, and ice thickness variables are used as output to construct a small-to-medium scale numerical model, WRF-Icing-Fog, which can directly perform rime ice accumulation calculations.

[0032] Preferably, the meteorological element data for each layer of the vertical layer of the Thompson-aero-MP cloud microphysics scheme calculation model includes: multi-layer radial wind U, zonal wind V, liquid water content Q, air temperature T, precipitation P, and air density M described by mathematical models, expressed by the following formulas (4) to (9):

[0033] U = w u(t，i，j，k) (4)

[0034] V = w v(t，i，j，k) (5)

[0035] Q = w q(t，i，j，k (6)

[0036] M = w m(t，i，j，k) (7)

[0037] T = w t(t，i，j，k) (8)

[0038] P = w p(t，i，j，k) (9)

[0039] In the formula:

[0040] t represents time, i, j, k represent the spatial coordinates of the small-to-medium scale model, and W u W v W q W m W t W p It is a function that describes the changes of six meteorological elements over time and space.

[0041] Preferably, the coupling of the MAKKONEN hoarfrost and ice accumulation model for calculating ice accumulation thickness includes:

[0042] The diameter D of the rime ice was calculated using the MAKKONEN rime ice model and expressed as follows (10):

[0043] D = w d(U,V,Q,M,T,P) (10)

[0044] In the formula:

[0045] W d The function representing the relationship between ice thickness and six four-dimensional meteorological elements calculated by the Thompson-aero-MP cloud microphysics scheme;

[0046] The diameter of the rime ice at each time step is calculated based on the ice accumulation diameter, and is expressed by the following formula (11):

[0047] D(t+Δt)=D(t)+ΔD (11)

[0048] In the formula:

[0049] Δt is the time step, and ΔD represents the ice accumulation increment within the time step Δt.

[0050] Preferably, the high-resolution numerical prediction model for rime ice accumulation in the generated area includes:

[0051] Using preprocessed USGS-annualized MODIS satellite 500m×500m resolution land cover data, we updated and replaced the less accurate and older land cover data in the small-to-medium scale numerical model WRF-Icing-Fog, which can directly perform rime ice calculations.

[0052] Preferably, the high-resolution numerical prediction model for rime ice accumulation in the generated area includes:

[0053] The small-scale numerical model WRF-Icing-Fog was compiled on the Rocky Linux system using the OneAPI / Intel compiler to generate a regional high-resolution numerical forecast model for rime ice and frost. The obtained reanalysis meteorological data or NCEP / GDAS data were used as the driving force, processed by the WRF-ARW preprocessing system WPS, and input into the regional high-resolution numerical forecast model for rime ice and frost for parallel computation.

[0054] Preferably, the prediction results are optimized by: analyzing the prediction results using the root mean square error (RMSE).

[0055]

[0056] Where D is the predicted value and O is the actual value;

[0057] If the RMSE value is less than the first threshold, the model's predictive accuracy is rated as good. If the RMSE is 0, the model's prediction is excellent. If the RMSE is greater than the first threshold, the quality of the data, the model's parameters, or the model structure should be checked.

[0058] Preferably, when constructing a high-resolution numerical prediction model for rime ice accumulation on the plateau, the vertical scheme of the model is set to Hybrid coordinates, the number of vertical layers is set to 51, and the terrain-following mass Eulerian coordinate system in WRF-ARW is replaced.

[0059] When establishing a high-resolution numerical prediction model for rime ice accumulation on the plateau, the data output dependency library is set to NETCDF4 to reduce the volume of the output results and the storage space occupied by the output data.

[0060] The second aspect of the present invention provides a hoarfrost and icing forecasting system that takes into account the underlying surface and cloud physical parameters, and the hoarfrost and icing forecasting method that takes into account the underlying surface and cloud physical parameters includes: a data preprocessing module, a numerical simulation / forecasting module, a model configuration module, and a result processing module;

[0061] The data preprocessing module acquires, preprocesses, and converts the data format and projection to make it suitable for small-scale mode reading in WRF-ARW.

[0062] The numerical simulation / forecasting module uses the WRF-ARW model and the MAKKONEN model to calculate multiple meteorological parameters and the thickness of rime ice accumulation;

[0063] The model configuration module configures and establishes a rime forecast model based on a specific region and resolution.

[0064] The ice accumulation result processing module processes the calculated results of rime ice diameter and generates rime ice-related products with high spatiotemporal resolution.

[0065] A third aspect of the present invention provides a high-performance computer device, including a memory and a processor, the memory storing a computer Fortran program, and the processor executing the computer Fortran program to perform the steps of the method described therein.

[0066] A fourth aspect of the present invention provides a computer-readable and writable storage medium having a computer Fortran program stored thereon, wherein the computer Fortran program, when executed by a processor, implements the steps of the rime ice prediction method taking into account the underlying surface and cloud physical parameterization.

[0067] Compared with the prior art, the beneficial effects of the present invention include at least the following:

[0068] (1) This invention combines the WRF-ARW mesoscale numerical weather prediction model and the MAKKONEN rime ice model. It fully couples the Thompson-aero-MP cloud microphysics parameterization scheme from the WRF-ARW model with the MAKKONEN rime ice model, producing historical rime ice diameter data with a 1-hour time interval and a spatial resolution of 2km×2km. Due to my country's complex terrain and numerous mountains, rime ice problems arise due to temperature drops caused by increasing altitude. Therefore, the model considers rime ice forecasting based on seasonality and terrain. To this end, the Thompson-aero-MP cloud microphysics parameterization scheme from the WRF-ARW mesoscale model and the MAKKONEN rime ice model are fully coupled to construct the WRF-Icing-Fog fully coupled rime ice numerical prediction model. This significantly reduces computational steps, allowing direct acquisition of rime ice diameter and rime ice quality detection or forecast data from the WRF-ARW model output.

[0069] (2) The hoarfrost data generated by the model dynamic downscaling based on the small-scale model WRF-ARW has higher spatiotemporal resolution (1 hour, 2km×2km), fully taking into account the complex underlying surface of the plateau. Therefore, it can obtain more detailed spatiotemporal distribution characteristics of plateau hoarfrost under complex underlying surface conditions.

[0070] (3) The WRF-Icing-Fog fully coupled numerical prediction model for rime ice accumulation takes into account the complex underlying surface effects, and provides a more accurate and realistic description of local rime ice accumulation data caused by complex vegetation cover on the plateau underlying surface. Attached Figure Description

[0071] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0072] in:

[0073] Figure 1 A flowchart illustrating a numerical prediction method for rime ice accumulation considering complex underlying surfaces, provided as an embodiment of the present invention;

[0074] Figure 2 This is a flowchart illustrating a numerical prediction system for rime ice accumulation considering complex underlying surfaces, provided as an embodiment of the present invention. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0076] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0077] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0078] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0079] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0080] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0081] like Figure 1 As shown, Embodiment 1 of the present invention provides a method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters, including the following steps:

[0082] Step 1: Collect high-resolution MODIS (Moderate Resolution Imaging Spectroradiometer) land cover data, stitch and preprocess it to obtain a high-precision underlying surface dataset.

[0083] In a preferred but non-limiting embodiment of the present invention, step 1 specifically includes:

[0084] Step 1.1: Obtain the annual MODIS satellite 500m × 500m resolution land cover dataset through USGS (US Geological Survey), and stitch it together to obtain the stitched high-resolution land cover data.

[0085] Step 1.2: Tools such as Python3, Geopandas, tiff, and GDAL can be used to preprocess the stitched high-resolution land cover data obtained in Step 1.1 and convert the data format to a binary format that can be read by WRF-ARW mode to obtain a high-precision underlying surface dataset.

[0086] The preprocessing includes projection transformation and resampling. The projection transformation is expressed by the following formula (1):

[0087] C(x, y) = LU(i, j) (1)

[0088] In the formula:

[0089] C represents the original land cover data;

[0090] (x, y) represents the location coordinates of the raster data after projection transformation;

[0091] LU represents the projection function;

[0092] (i, j) represents the location coordinates of the raster data before projection transformation.

[0093] The original land cover data C is resampled to a new raster using the four-point interpolation function IN, as shown in the following formula (2):

[0094] C′(m,n)=IN[C(x,y)] (2)

[0095] In the formula:

[0096] C(x, y) represents the raw land cover data for location (x, y);

[0097] IN represents the four-point interpolation function;

[0098] C′(m, n) represents the preprocessed land cover raster data;

[0099] The m×n raster matrix is ​​converted into a binary file A that can be recognized by WRF-ARW mode using the conversion function Tr, as shown in the following formula (3):

[0100] A = Tr(C′) (3)

[0101] In the formula:

[0102] A indicates a binary file that the WRF-ARW mode can recognize;

[0103] Tr represents the conversion function used to convert binary files that can be recognized by WRF-ARW mode.

[0104] Step 2: Integrate the MAKKONEN hoarfrost and ice accumulation model into the Thompson-aero-MP cloud microphysical parameterization scheme of the WRF-ARW model to obtain the small-to-medium scale numerical model WRF-Icing-Fog, which can directly perform hoarfrost and ice accumulation calculations.

[0105] Preferably, but not limitingly, step 2 specifically includes:

[0106] Step 2.1: Based on the Thompson-aero-MP cloud microphysics scheme in WRF-ARW, calculate meteorological data such as wind direction, wind speed, liquid water content, air density, temperature and precipitation for each layer of the vertical layer of the model.

[0107] Specifically, the variables in the small-scale numerical model WRF-ARW used for icing calculation include: multi-level radial wind U, zonal wind V, liquid water content Q, air temperature T, precipitation P, and air density M, described by mathematical models, as expressed in the following formulas (4) to (9):

[0108] U = w u(t，i，j，k) (4)

[0109] V = w v(t，i，j，k) (5)

[0110] Q = w q(t，i，j，k) (6)

[0111] M = w m(t，i，j，k) (7)

[0112] T = w t(t，i，j，k) (8)

[0113] P = w p(t，i，j，k) (9)

[0114] In the formula:

[0115] t represents time, i, j, k represent the spatial coordinates of the small-to-medium scale model, and W u W v W q W m W t W p It is a function that describes the changes of six meteorological elements over time and space.

[0116] Step 2.2: Integrate the MAKKONEN rime ice model into the Thompson-aero-MP cloud microphysics parameterization scheme of the WRF-ARW model. This includes: using the six meteorological elements provided by the Thompson-aero-MP cloud microphysics scheme of the WRF-ARW model as inputs for calculating the rime ice thickness of the MAKKONEN rime ice model; calculating the rime ice thickness at each time step according to the rime ice thickness calculation formula in the MAKKONEN rime ice model; and outputting the rime ice thickness variable.

[0117] Specifically, the diameter D of the rime ice is calculated using the MAKKONEN rime ice model, and is expressed by the following formula (10):

[0118] D = w d(U,V,Q，M,T,P) (10)

[0119] In the formula:

[0120] W d This represents the relationship function between ice thickness and the six four-dimensional meteorological elements calculated by the WRF-ARW cloud microphysics parameterization scheme. Step 2.3: Calculate the diameter of the rime ice at each time step based on the ice diameter, expressed by the following formula (11):

[0121] D(t+Δt)=D(t)+ΔD (11)

[0122] In the formula:

[0123] Δt is the time step, and ΔD represents the ice accumulation increment within the time step Δt.

[0124] Preferably, but not restrictively, the Lambert projection is used when establishing the meteorological model; vertical layers: both nested layers are 51 layers, of which 26 layers are below 1km vertically, using a hybrid coordinate system to reduce the impact of gravity waves from high plateaus and mountains on the upper-air meteorological field and its impact on the surrounding and downstream weather; the spatial resolution of the nested inner layer is 2km×2km, and the time step is 60 seconds; the cloud microphysics scheme is Thompson-aero-MP, the boundary layer transmission scheme is MYNN 2.5, the near-surface layer scheme is MYNN 2.5, the shortwave radiation scheme is RRTMG, the longwave radiation scheme is RRTMG, the land surface parameterization scheme is Noah-MP, the cloud droplet / fog droplet gravity sedimentation scheme is enabled, and the topographic surface wind correction scheme Topo_Wind is enabled. The detailed selection of each parameterization scheme is shown in Table 1 below, and the calculation results are output hourly.

[0125] Table 1. Combinations of physical schemes for the WRF-Icing-Fog model

[0126] Microphysics scheme Thompson-aero-MP Longwave radiation scheme RRTMG Shortwave radiation scheme RRTMG Boundary layer scheme MYNN 2.5 Land surface scheme Noah-MP Near-surface scheme MYNN 2.5 Cumulus convection parameterization scheme GF (Inner Layer Closed) Topographic and surface wind correction scheme Open Vertical coordinate system Hybrid Topo_Wind Open grav_settling Open

[0127] Step 3: Collect atmospheric reanalysis meteorological data or forecasting system data as required by the small-scale numerical model WRF-Icing-Fog.

[0128] Preferred but not limited, it is advisable to obtain and download common global atmospheric reanalysis meteorological data or global forecast system data from WMO or domestic and international sources (such as Europe and the United States or China): download publicly available global atmospheric reanalysis meteorological data or global forecast system data, such as ERA5, NCEP GFS (Global Forecast System), NCEP FNL, etc., from the WMO World Meteorological Centre website or the websites of European and American meteorological departments.

[0129] The meteorological data driving the WRF-ARW model must include: U, V, air pressure, temperature, humidity (with no less than 30 vertical layers), ground 0cm temperature, soil moisture, soil temperature, snow depth, snow cover, and other information to drive the fully coupled hoarfrost and icing numerical forecasting model WRF-Icing-Fog.

[0130] Step 4: Replace the original static underlying surface data of the WRF-ARW model with the high-precision underlying surface dataset obtained in Step 1, and then process the atmospheric reanalysis meteorological data or forecast system data obtained in Step 3 through the WRF-ARW preprocessing system WPS to compile and generate a regional high-resolution numerical forecast model for rime ice accumulation, thereby obtaining regional, high spatiotemporal resolution rime ice accumulation simulation or prediction results that consider complex underlying surfaces.

[0131] Preferably, but not limitingly, step 4 specifically includes:

[0132] Step 4.1: Obtain the underlying surface static dataset provided by the small-scale model WRF-ARW, which contains insufficiently accurate and older underlying land cover. Use the USGS annual value MODIS satellite 500m×500m resolution land cover data preprocessed in Step 1 to update and replace the insufficiently accurate and older land cover data in the small-scale numerical model WRF-Icing-Fog, which can be directly used for rime ice calculation, thereby enabling calculations based on higher resolution and higher accuracy land use data.

[0133] Step 4.2: Use the WRF-ARW preprocessing system WPS to set and design the configuration to obtain the simulation or forecast area configuration with the specified study or area of ​​interest range and resolution, and generate the static underlying surface data file of the regional high-resolution hoarfrost numerical prediction model: geo_em.d0X.nc, for subsequent reading by the generated regional high-resolution hoarfrost numerical prediction model.

[0134] Step 4.3: Use the OneAPI / Intel compiler to compile the small-scale numerical model WRF-Icing-Fog on the Rocky Linux system to generate a regional high-resolution numerical prediction model for rime ice accumulation.

[0135] Step 4.4: Using the obtained reanalysis meteorological data or NCEP / GDAS data as the driving force, the data is processed by the WRF-ARW preprocessing system WPS, and input into the regional high-resolution hoarfrost numerical prediction model for large-scale parallel computation. This yields high spatiotemporal resolution hoarfrost diameter prediction results for the covered area, taking into account the complex underlying surface of the plateau, and the prediction results are then improved.

[0136] When constructing a high-resolution numerical prediction model for rime ice on the plateau, the vertical scheme of the model is set to Hybrid coordinates, the number of vertical layers is set to 51, and the terrain-following mass Eulerian coordinate system in WRF-ARW is replaced. When establishing a high-resolution numerical prediction model for rime ice on the plateau, the WRF-Icing-Fog model data output dependency library is set to NETCDF4 (supported by hdf5) to reduce the size of the output results and reduce the storage space occupied by the output data.

[0137] Preferably, but not limitingly, optimizing the prediction results includes analyzing the prediction results using the root mean square error (RMSE), expressed as follows:

[0138]

[0139] In the formula:

[0140] D represents the predicted value, and O represents the actual value.

[0141] If the RMSE value is less than the first threshold, the model's predictive accuracy is rated as good. If the RMSE is 0, the model's prediction is excellent. If the RMSE is greater than the first threshold, the quality of the data, the model's parameters, or the model structure should be checked.

[0142] refer to Figure 2 Embodiment 2 of the present invention provides a hoarfrost and icing forecasting system that takes into account the underlying surface and cloud physical parameters, and runs the hoarfrost and icing forecasting method described in Embodiment 1 that takes into account the underlying surface and cloud physical parameters. The system includes: a data preprocessing module, a numerical simulation / forecasting module, a model configuration module, and a result processing module; wherein,

[0143] The data preprocessing module acquires, preprocesses, and converts the data format and projection to make it suitable for small-scale mode reading in WRF-ARW.

[0144] The numerical simulation / forecasting module uses the WRF-ARW model and the MAKKONEN model to calculate six meteorological parameters and the thickness of rime ice accumulation;

[0145] The model configuration module configures and establishes a rime forecast model based on a specific region and resolution.

[0146] The ice accumulation result processing module processes the calculated results of rime ice diameter and generates rime ice-related products with high spatiotemporal resolution.

[0147] Embodiment 3 of the present invention provides a high-performance computer device, including a memory and a processor. The memory stores a computer Fortran program, and the processor executes the computer Fortran program to implement the steps of the rime ice prediction method that takes into account the underlying surface and cloud physical parameters as described in Embodiment 1.

[0148] Embodiment 4 of the present invention provides a computer-readable and writable storage medium storing a computer Fortran program, which, when executed by a processor, implements the steps of the rime ice prediction method that takes into account the underlying surface and cloud physical parameters as described in Embodiment 1.

[0149] It is understood that in embodiments 3 and 4 of the present invention, if the functions described are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0151] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0152] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0153] As can be understood from the above description, the present invention has outstanding substantive features and significant progress. In particular, compared with the prior art documents 1 and 2, it includes at least the following: The present invention directly couples the icing model to the cloud microphysical parameterization scheme of the small and medium scale model, realizing the direct output of icing models at different height levels. It is not necessary to calculate and output meteorological elements from the small and medium scale model and then input them into the icing model. It can also output icing prediction results at different vertical heights at the same time.

[0154] To more clearly illustrate the outstanding substantive features of the technical solution of this invention and the significant progress it brings to the prior art, the following provides a verification example of the numerical prediction method for rime ice accumulation considering complex underlying surfaces.

[0155] The method for predicting rime ice accumulation that takes into account the underlying surface and cloud physical parameters in the validation example includes the following steps:

[0156] Step 1: Based on the USGS, obtain the annual MODIS satellite 500m×500 resolution underlying surface dataset and construct a high-precision underlying surface dataset covering the plateau that can be identified and utilized by WRF-ARW.

[0157] Step 2: Based on the cloud microphysical parameterization scheme in the WRF-ARW model—the Thompson-aero-MP scheme and the MAKKONEN hoarfrost model, establish a hoarfrost numerical prediction model WRF-Icing-Fog that fully couples the WRF-ARW and MAKKONEN hoarfrost models;

[0158] Step 3: Obtain and download WMO or other commonly used global atmospheric reanalysis meteorological data or global forecasting system data: Download publicly available global atmospheric reanalysis meteorological data or global forecasting system data, such as ERA5, NCEP GFS, NCEP FNL, etc., from the WMO World Meteorological Centre website or the websites of European and American meteorological departments. The meteorological data driving the WRF-ARW model must include: U, V, air pressure, temperature, humidity (at least 30 vertical layers), surface 0cm temperature, soil moisture, soil temperature, snow depth, snow cover, etc., to drive the fully coupled hoarfrost and icing numerical prediction model WRF-Icing-Fog;

[0159] Step 4: Input the underlying surface and global atmospheric reanalysis meteorological data or global forecast system data obtained in Step 1 and Step 3 into the regional high-resolution hoarfrost numerical prediction model WRF-Icing-Fog for large-scale parallel computation to obtain regional hoarfrost products with high spatiotemporal resolution that take into account the complex underlying surface of the plateau, including icing time, icing space and icing diameter. When establishing the meteorological model, Lambert projection was used; vertical layers: both nested layers have 51 layers, of which 26 layers are below 1km vertically. A hybrid coordinate system was used to reduce the impact of gravity waves from high plateaus and mountains on the upper-air meteorological field and its impact on the surrounding and downstream weather. The spatial resolution of the nested inner layer is 2km×2km, and the time step is 60 seconds. The cloud microphysics scheme was selected as Thompson-aero-MP, the boundary layer transmission scheme was MYNN 2.5, the near-surface layer scheme was MYNN 2.5, the shortwave radiation scheme was RRTMG, the longwave radiation scheme was RRTMG, the land surface parameterization scheme was Noah-MP, the cloud droplet / fog droplet gravity sedimentation scheme was enabled, and the topographic surface wind correction scheme Topo_Wind was enabled. The detailed selection of each parameterization scheme is shown in Table 1 below. The calculation results are output hourly.

[0160] By inputting 20 years of global atmospheric reanalysis meteorological data from NCEP FNL into the WRF-Icing-Fog model for large-scale parallel computation, a rime ice dataset with a regional spatial resolution of 2 km × 2 km and hourly data was obtained.

[0161] The meanings of the abbreviations used in this invention are explained as follows:

[0162] WMO: World Meteorological Organization;

[0163] ECMWF: European Centre for Medium-Range Weather Forecasts;

[0164] NCEP: National Centers for Environmental Prediction, United States.

[0165] GDAS: Global Data Assimilation System;

[0166] MODIS: Moderate Resolution Imaging Spectroradiometer;

[0167] USGS: US Geological Survey;

[0168] NOAA: National Oceanic and Atmospheric Administration.

Claims

1. A method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters, characterized in that: Includes the following steps: High-resolution MODIS land cover data were collected, stitched together, and preprocessed to obtain a high-precision underlying surface dataset. By fully coupling the MAKKONEN hoarfrost and ice accumulation model with the cloud microphysical parameterization scheme in the WRF-ARW model, a small-to-medium scale numerical model WRF-Icing-Fog, which can directly perform hoarfrost and ice accumulation calculations, is obtained. Collect atmospheric reanalysis meteorological data or forecasting system data in accordance with the data requirements of the small- and medium-scale numerical model WRF-Icing-Fog. The original static underlying surface data of the WRF-ARW model is replaced with the high-precision underlying surface dataset. Then, the atmospheric reanalysis meteorological data or forecast system data are processed by the WRF-ARW preprocessing system WPS to compile and generate a regional high-resolution numerical forecast model for rime ice and icing. This results in regional, high spatiotemporal resolution simulation or prediction results for rime ice and icing that take into account complex underlying surfaces.

2. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters as described in claim 1, characterized in that: The preprocessing includes: projection transformation, resampling, and binary format conversion. The projection transformation is expressed by the following formula (1): C(x,y)=LU(i,j) (1) In the formula: C represents the original land cover data, (x,y) represents the location coordinates of the raster data after projection transformation, LU represents the projection function, and (i,j) represents the location coordinates of the raster data before projection transformation. The original land cover data C is resampled to a new raster using the four-point interpolation function IN, as shown in the following formula (2): C'(m,n)=IN[C(x,y)] (2) In the formula: C(x,y) represents the original land cover data at location (x,y), IN represents the four-point interpolation function, and C'(m,n) represents the resampled land cover raster data; The m×n raster matrix is ​​converted into a WRF-ARW pattern recognition binary file A using the transformation function Tr, as shown in the following formula (3): A = Tr(C') (3) In the formula: A represents the binary file that WRF-ARW mode can recognize, and Tr represents the conversion function used to convert it into a binary file that WRF-ARW mode can recognize.

3. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 1, characterized in that: By fully coupling the MAKKONEN hoarfrost and icing model with the cloud microphysical parameterization scheme in the WRF-ARW model, a small-to-medium scale numerical model, WRF-Icing-Fog, capable of directly calculating hoarfrost and icing, is obtained, including: Meteorological element data for each layer of the vertical layer of the computational model based on the Thompson-aero-MP cloud microphysics scheme in WRF-ARW; The MAKKONEN rime ice accumulation model is coupled to calculate ice thickness. Meteorological data for each layer are used as input to the MAKKONEN rime ice accumulation model, and ice thickness variables are used as output to construct a small-to-medium scale numerical model, WRF-Icing-Fog, which can directly perform rime ice accumulation calculations.

4. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 3, characterized in that: The meteorological data for each layer of the Thompson-aero-MP cloud microphysics scheme calculation model vertical layer includes: multi-layer radial wind U, zonal wind V, liquid water content Q, air temperature T, precipitation P, and air density M, described by mathematical models and expressed by the following formulas (4) to (9): U=w u(t,i,j,k) (4) V=w v(t,i,j,k) (5) Q=w q(t,i,j,k) (6) M=w m(t,i,j,k) (7) T=w t(t,i,j,k) (8) P=w p(t,i,j,k) (9) In the formula: t represents time, i, j, k represent the spatial coordinates of the small-to-medium scale model, and W u W v W q W m W t W p It is a function that describes the changes of six meteorological elements over time and space.

5. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 4, characterized in that: The coupled MAKKONEN hoarfrost and ice accumulation model is used to calculate ice accumulation thickness, including: The diameter D of the rime ice was calculated using the MAKKONEN rime ice model and expressed as follows (10): D=w d(U,V,Q,M,T,P) (10) In the formula: w d The function representing the relationship between ice thickness and six four-dimensional meteorological elements calculated by the Thompson-aero-MP cloud microphysics scheme; The diameter of the rime ice at each time step is calculated based on the ice accumulation diameter, and is expressed by the following formula (11): D(t+Δt)=D(t)+ΔD (11) In the formula: Δt is the time step, and ΔD represents the ice accumulation increment within the time step Δt.

6. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 1, characterized in that: The high-resolution numerical prediction model for rime ice accumulation in the generated area includes: Using preprocessed USGS-annualized MODIS satellite 500m×500m resolution land cover data, we updated and replaced the less accurate and older land cover data in the small-to-medium scale numerical model WRF-Icing-Fog, which can directly perform rime ice calculations.

7. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 1 or 6, characterized in that: The high-resolution numerical prediction model for rime ice accumulation in the generated area includes: The small-scale numerical model WRF-Icing-Fog was compiled on the Rocky Linux system using the OneAPI / Intel compiler to generate a regional high-resolution numerical forecast model for rime ice and frost. The obtained reanalysis meteorological data or NCEP / GDAS data were used as the driving force, processed by the WRF-ARW preprocessing system WPS, and input into the regional high-resolution numerical forecast model for rime ice and frost for parallel computation.

8. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 1, characterized in that: The prediction results are optimized, including by analyzing the prediction results using the root mean square error (RMSE). Where D is the predicted value and O is the actual value; If the RMSE value is less than the first threshold, the model's predictive accuracy is rated as good. If the RMSE is 0, the model's prediction is excellent. If the RMSE is greater than the first threshold, the quality of the data, the model's parameters, or the model structure should be checked.

9. The method for forecasting rime ice accumulation that takes into account the underlying surface and cloud physical parameters according to claim 1, characterized in that: When constructing a high-resolution numerical prediction model for rime ice accumulation on the plateau, the vertical scheme of the model is set to Hybrid coordinates, the number of vertical layers is set to 51, and the terrain-following mass Eulerian coordinate system in WRF-ARW is replaced. When establishing a high-resolution numerical prediction model for rime ice accumulation on the plateau, the data output dependency library is set to NETCDF4 to reduce the volume of the output results and the storage space occupied by the output data.

10. A hoarfrost and icing forecasting system that takes into account the underlying surface and cloud physical parameters, operating the hoarfrost and icing forecasting method that takes into account the underlying surface and cloud physical parameters as described in any one of claims 1-9, characterized in that, include: Data preprocessing module, numerical simulation / forecasting module, model configuration module, results processing module; The data preprocessing module acquires, preprocesses, and converts the data format and projection to make it suitable for small-scale mode reading in WRF-ARW. The numerical simulation / forecasting module uses the WRF-ARW model and the MAKKONEN model to calculate multiple meteorological parameters and the thickness of rime ice accumulation; The model configuration module configures and establishes a rime forecast model based on a specific region and resolution. The ice accumulation result processing module processes the calculated results of rime ice diameter and generates rime ice-related products with high spatiotemporal resolution.

11. A high-performance computer device, comprising a memory and a processor, wherein the memory stores computer Fortran programs, characterized in that, When the processor executes the computer Fortran program, it implements the steps of the rime ice prediction method according to any one of claims 1 to 9, which takes into account the underlying surface and cloud physical parameters.

12. A computer-readable and writable storage medium storing a computer Fortran program thereon, characterized in that, When the computer Fortran program is executed by the processor, it implements the steps of the rime ice prediction method according to any one of claims 1 to 9, which takes into account the underlying surface and cloud physical parameterization.

Images