A hundred-meter-level terrain processing method for preserving mode terrain refinement features

Abstract

The application discloses a kind of hundred-meter level terrain processing methods of reserved mode topography fine feature, it is related to numerical weather prediction technical field, including, according to topography spectrum constraint topography set, the change of topography power spectrum under the action of topography smoothing function is analyzed, and topography smoothing function is adjusted, and improved smoothing function configuration data are generated;According to improved smoothing function configuration data, hundred-meter level basic topography set is smoothed, and spectrum fidelity smoothing topography data are generated;Spectrum fidelity smoothing topography data are carried out topography power spectrum consistency check and topography height deviation check, and spectrum fidelity mode topography field is generated;Spectrum fidelity mode topography field is written in numerical weather prediction mode pre-processing interface, and spectrum fidelity pre-processing topography set is generated.The application achieves the effect of improving mode topography fine expression ability, spectrum consistency and numerical weather prediction mode pre-processing adaptability.

Description

Technical Field

[0001] This invention relates to the field of numerical weather prediction technology, and in particular to a method for topographic processing at the 100-meter level that preserves the refined features of the model topography. Background Technology

[0002] As the resolution of regional numerical weather prediction models continues to improve, topographic data at the 100-meter level is gradually entering the model preprocessing workflow. Projection transformation, grid mapping, and multi-nested domain adaptation of topographic elevation fields have become important foundations for the construction of high-precision models. Analysis methods based on topographic power spectrum characteristics are also increasingly being applied to assess the quality of model topographic representations, focusing on the impact of complex terrain on boundary layer structure, local uplift processes, and small-to-medium-scale weather evolution.

[0003] Existing technologies typically rely on a uniform terrain smoothing function to improve integral stability when incorporating high-resolution terrain into model preprocessing. However, the smoothing process can weaken the terrain power spectrum components in the key wavelength range, making it difficult for the model terrain field to simultaneously maintain fine terrain features and spectral consistency at the resolution level, thereby affecting the accuracy of subsequent model terrain input. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a 100-meter-level terrain processing method that preserves the refined features of the model terrain, solving the problem that the fine spectral features of 100-meter-level terrain are difficult to maintain during the model preprocessing smoothing process.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a 100-meter-level terrain processing method that preserves the refined features of model terrain. The method includes: acquiring 100-meter-level raw terrain data and model grid parameters, performing unified projection transformation and grid mapping to generate a 100-meter-level basic terrain set; extracting terrain power spectrum features at different spatial scales from the 100-meter-level basic terrain set, and configuring corresponding terrain spectrum preservation requirements according to the resolution hierarchy of the model grid parameters to generate a terrain spectrum-constrained terrain set; analyzing the changes in terrain power spectrum under the action of a terrain smoothing function based on the terrain spectrum-constrained terrain set, and adjusting the terrain smoothing function to generate improved smoothing function configuration data; smoothing the 100-meter-level basic terrain set based on the improved smoothing function configuration data to generate spectrum-fidelity smoothed terrain data; performing terrain power spectrum consistency verification and terrain height deviation verification on the spectrum-fidelity smoothed terrain data to generate a spectrum-fidelity model terrain field; and writing the spectrum-fidelity model terrain field into the numerical weather prediction model preprocessing interface to generate a spectrum-fidelity preprocessed terrain set.

[0007] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined terrain features of the present invention, the hundred-meter-level original terrain data includes elevation data, coordinate location data, regional range data, and terrain grid distribution information. The pattern grid parameters include the nesting hierarchy between parent and child domains, the grid resolution of each domain, the grid range, the grid number, the projection parameters, and the corresponding relationship of the pattern regions; The terrain smoothing function is the terrain smoothing function in the preprocessing program of the numerical weather prediction model. The numerical weather prediction model preprocessing interface is generated by the numerical weather prediction model preprocessing program based on the nested hierarchy of parent and child domains, the grid resolution of each domain, the grid range, and the projection parameters in the model grid parameters.

[0008] As a preferred embodiment of the 100-meter-level terrain processing method for preserving the refined features of the terrain pattern described in this invention, the generation of the 100-meter-level basic terrain set specifically includes: The raw terrain data at the 100-meter level is transformed using a unified projection datum to generate unified projection terrain data. Based on the grid range, grid number and pattern region correspondence in the model grid parameters, the unified projection terrain data is divided into domains, labeled with regional affiliation and raster position correspondence, and generated into domain-mapped terrain data. Based on the nested hierarchical relationship between parent and child domains in the model grid parameters, the grid resolution and projection parameters of each domain, cross-domain position alignment, resolution adaptation and hierarchical association are performed on the domain-mapped terrain data to generate a 100-meter-level basic terrain set.

[0009] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined terrain features of the present invention, the extraction of terrain power spectrum features at different spatial scales specifically includes: The topographic elevation distribution of each domain is extracted from the basic topographic data at the 100-meter level, and the analysis scale is set according to the grid resolution of each domain at different spatial scales to generate multi-scale topographic data of different domains. Scale separation and power spectrum analysis of terrain elevation undulation within each spatial analysis window of multi-scale terrain data in the domain are performed to generate frequency band response data. Based on the nested hierarchical relationship between parent and child domains, the frequency band response data is registered and scaled between different domain layers to extract the topographic power spectrum features at different spatial scales.

[0010] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined features of the terrain pattern described in this invention, the generation of the terrain spectrum-constrained terrain set specifically includes, The topographic power spectrum features at different spatial scales are classified by scale and hierarchically mapped to generate hierarchical frequency band identification data. Based on the resolution level of the model grid parameters, the corresponding topographic spectrum retention requirements are configured for different wavelength intervals in the layered frequency band identification data to generate spectrum constraint data. Based on the terrain location of each domain in the 100-meter-level basic terrain set, the spectral constraint data is divided into domains and encapsulated to generate a terrain spectral constraint terrain set.

[0011] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined terrain features of the present invention, the analysis of terrain power spectrum changes under the action of the terrain smoothing function specifically includes, Based on the nested hierarchical relationship between parent and child domains and the correspondence between pattern regions, spectral constraint information is extracted from the topographic spectral constraint terrain set, and then divided into domains and normalized to generate regional spectral constraint data. Based on the regional spectral constraint data, intensity and wavelength distribution comparisons are performed on the topographic power spectrum under the action of the topographic smoothing function to generate regional power spectrum comparison data. Based on regional power spectrum comparison data, the differences in wavelength and intensity distribution of the topographic power spectrum under the action of the topographic smoothing function are identified, and regional power spectrum difference data are generated.

[0012] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined features of the terrain pattern described in this invention, the generation of improved smoothing function configuration data specifically includes: The regional power spectrum difference data is segmented and labeled to generate smooth difference segment data; Based on the smoothing difference segmented data, the weight configuration of the central grid point and adjacent grid points within the terrain smoothing function is adjusted to generate smoothing function weight data; Based on the terrain set constrained by the topographic wave spectrum, the smoothing function weight data is divided into domains and integrated to generate improved smoothing function configuration data.

[0013] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined features of the terrain pattern described in this invention, the generation of spectral-fidelity smoothed terrain data specifically includes: Based on the improved smoothing function configuration data, the elevation data of the 100-meter-level basic terrain set is smoothed to generate initial smoothed terrain data. Based on the nested hierarchical relationship between parent and child domains and the grid resolution of each domain in the pattern grid parameters, the initial smoothed terrain data is divided into domains and aligned with the location to generate domain smoothed terrain data. Based on the topographic spectrum constrained topographic set, the domain-smoothed topographic data is integrated to generate spectrally faithful smoothed topographic data.

[0014] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined features of the pattern terrain described in this invention, the generation of the spectral fidelity pattern terrain field specifically includes: A domain-by-domain topographic power spectrum comparison was performed between the spectral-fidelity smoothed topographic data and the 100-meter-level basic topographic set to generate domain-by-domain spectral comparison data. Based on the topographic spectral constraint topographic set, the consistency of the domain-by-domain spectral comparison data is verified to generate spectral consistency data. Based on the spectral consistency data and the elevation data of the 100-meter-level basic terrain set, the terrain height deviation of the spectral fidelity smoothed terrain data is checked to generate spectral checked terrain data. Cross-domain consistency verification and merging are performed on the spectral verification terrain data to generate a spectral fidelity model terrain field.

[0015] As a preferred embodiment of the hundred-meter-level terrain processing method for preserving the refined features of the terrain pattern described in this invention, the generation of the spectral fidelity preprocessed terrain set specifically includes: Based on the numerical weather prediction model preprocessing interface, the topographic field of the spectral fidelity model is meshed and restructured to generate preprocessed adapted topographic data. Based on the nested hierarchical relationship between parent and child domains, the terrain set constrained by the terrain spectrum, and the configuration data of the improved smoothing function, the pre-processed adaptive terrain data is checked for consistency before being written, and pre-processed verification terrain data is generated. Based on the numerical weather prediction model preprocessing interface, the preprocessed verification terrain data is organized by domain and written into the numerical weather prediction model preprocessing interface to generate a spectral fidelity preprocessed terrain set.

[0016] The beneficial effects of this invention are as follows: by analyzing the changes in the topographic power spectrum under the action of the topographic smoothing function based on the topographic spectral constraints of the topographic set and generating improved smoothing function configuration data, the topographic smoothing function is used to differentiate the adjustment of different topographic spectral components. This makes the smoothing process consistent with the spectral retention requirements corresponding to the model grid parameters, which is used to generate spectrally accurate smoothed topographic data and support the construction of spectrally accurate model topographic fields. This achieves the effect of improving the model's ability to express topographic details, spectral consistency, and adaptability to numerical weather prediction model preprocessing. Attached Figure Description

[0017] 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.

[0018] Figure 1 A flowchart of a 100-meter-level terrain processing method to preserve the refined features of the model terrain.

[0019] Figure 2 A flowchart for generating a 100-meter-level basic terrain dataset.

[0020] Figure 3 A flowchart for generating terrain sets constrained by topographic wave spectrum.

[0021] Figure 4 A flowchart for configuring data generation to improve the smoothing function. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] 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.

[0024] 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.

[0025] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a method for 100-meter-level terrain processing that preserves refined features of the terrain pattern, comprising the following steps: S1. Collect raw terrain data and model grid parameters at the 100-meter level, and perform unified projection transformation and grid mapping processing to generate a basic terrain set at the 100-meter level.

[0026] S1.1, The raw terrain data at the 100-meter level includes elevation data, coordinate location data, regional range data, and terrain raster distribution information.

[0027] Specifically, terrain raster data is retrieved according to the spatial range corresponding to the target area, and the elevation and coordinate data of each terrain grid point are obtained one by one; the retrieved results are spatially truncated according to the boundary of the target area to form regional range data; the terrain grid points are organized according to the grid arrangement relationship within the target area to form terrain raster distribution information; the elevation data, coordinate data, regional range data, and terrain raster distribution information are linked and organized according to the correspondence of the same terrain grid point to obtain raw terrain data at the hundred-meter level.

[0028] S1.2 The model grid parameters include the nesting hierarchy between parent and child domains, the grid resolution of each domain, the grid range, the grid number, the projection parameters, and the corresponding relationship of the model regions.

[0029] Specifically, the hierarchical nesting settings of each domain are read from the grid configuration file of the numerical weather prediction model to form a nested hierarchical relationship between parent and child domains; the grid resolution of each domain is extracted based on the spatial discretization settings of each domain in the grid configuration file, and the grid extent is formed according to the boundary position of each domain within the model coverage area; the grid number is formed according to the identification order of each domain in the hierarchical structure, and the projection parameters are formed in combination with the map projection configuration; the model area correspondence is formed based on the correspondence between the grid extent of each domain and the spatial position of the target area; the nested hierarchical relationship between parent and child domains, the grid resolution of each domain, the grid extent, the grid number, the projection parameters, and the model area correspondence are associated and organized to obtain the model grid parameters.

[0030] S1.3. Perform unified projection datum conversion on the raw terrain data at the hundred-meter level to generate unified projection terrain data.

[0031] Specifically, the coordinate position data and regional range data of the raw terrain data at the 100-meter level are read, and the coordinate reference corresponding to the projection parameters in the model grid parameters is used as a unified projection reference. The coordinate position data of each terrain grid point is converted point by point. During the coordinate conversion process, the correspondence between the elevation data of each terrain grid point and the terrain grid distribution information is kept unchanged. The regional range data and terrain grid distribution information are reorganized according to the converted coordinate positions, so that the converted coordinate position data, elevation data, regional range data and terrain grid distribution information are under the same projection reference, thus generating unified projected terrain data.

[0032] S1.4. Based on the grid range, grid number and pattern region correspondence in the model grid parameters, perform domain clipping, region attribution marking and raster position correspondence on the unified projection terrain data to generate domain-mapped terrain data.

[0033] Specifically, using the grid range in the model grid parameters as the spatial boundary, the regional range data and coordinate position data in the unified projection terrain data are filtered domain by domain, and the elevation data and terrain raster distribution information falling within each grid range are extracted separately. According to the grid number in the model grid parameters, the extracted elevation data, coordinate position data, regional range data, and terrain raster distribution information within each grid range are labeled with the corresponding grid number to form a regional attribution mark consistent with the grid number. According to the model regional correspondence, the coordinate position data with the completed regional attribution mark is matched point by point with the raster position within the corresponding grid range, so that the elevation data, coordinate position data, regional range data, and terrain raster distribution information form a one-to-one correspondence within the corresponding grid range, generating domain-mapping terrain data.

[0034] S1.5. Based on the nested hierarchical relationship between parent and child domains in the model grid parameters, the grid resolution and projection parameters of each domain, perform cross-domain position alignment, resolution adaptation and hierarchical association on the domain-mapped terrain data to generate a 100-meter-level basic terrain set.

[0035] Specifically, based on the projection parameters in the model grid parameters, the projection reference of the coordinate position data of each domain in the domain-mapping terrain data is consistently corrected, so that the parent domain and the child domain form comparable spatial positions under the same projection expression; according to the nesting hierarchy of the parent domain and the child domain in the model grid parameters, the overlapping positions of the spatial coverage area corresponding to the parent domain grid in the target area and the spatial coverage area of ​​the child domain grid under the nesting setting after further refinement within the coverage area of ​​the parent domain are hierarchically matched, and the elevation data, coordinate position data and terrain raster distribution information corresponding to the overlapping positions are organized according to the same spatial position to complete the cross-domain position alignment; according to the grid resolution of each domain, the domain-mapping terrain data that has completed the cross-domain position alignment is position-aligned and raster-adapted according to the corresponding grid scale to complete the resolution adaptation; according to the nesting hierarchy of the parent domain and the child domain, the domain data that has completed the resolution adaptation are hierarchically associated to obtain a 100-meter-level basic terrain set.

[0036] S2. Extract the topographic power spectrum features at different spatial scales from the 100-meter-level basic topographic set, and generate a topographic spectrum constrained topographic set according to the resolution level configuration of the model grid parameters and the corresponding topographic spectrum retention requirements.

[0037] S2.1 Extract the terrain elevation distribution corresponding to each domain from the 100-meter-level basic terrain set, and set the analysis scale under different spatial scales according to the grid resolution of each domain to generate multi-scale terrain data of different domains.

[0038] Specifically, elevation and coordinate data are extracted from the 100-meter-level basic terrain dataset according to the boundaries of each domain. The elevation data are then arranged in spatial order based on the terrain raster distribution information of each domain, forming the terrain elevation distribution corresponding to each domain. Based on the grid resolution of each domain, the spatial scale corresponding to the grid unit of each domain is calculated, and multiple analysis scales corresponding to the grid unit scale are set around the terrain elevation distribution corresponding to each domain. The terrain elevation distribution corresponding to each domain is then organized hierarchically according to each analysis scale, so that the terrain elevation distribution under different analysis scales maintains the correspondence between the coordinate data and the terrain raster distribution information of each domain, generating multi-scale terrain data for each domain.

[0039] S2.2 Perform scale separation and topographic power spectrum analysis on the topographic elevation undulations in each spatial analysis window of the multi-scale topographic data to generate frequency band response data.

[0040] Specifically, based on the spatial range corresponding to each spatial analysis window, the continuously arranged terrain elevation distribution within the window is extracted from the multi-scale terrain data of the sub-domain, and the mean of the terrain elevation distribution within the window is removed to obtain the elevation undulation sequence within the corresponding window. According to the analysis scale corresponding to each spatial analysis window, the elevation undulation sequence is subjected to discrete spectrum transformation to obtain the spectral amplitude value of the corresponding wavenumber position. The power value corresponding to each wavenumber position is calculated based on the spectral amplitude value, and the wavenumber position is converted into a wavelength position in combination with the grid resolution of each domain. The power value corresponding to each wavelength position is accumulated in segments according to the preset wavelength interval to obtain the frequency band power distribution results corresponding to different wavelength intervals within each spatial analysis window. The frequency band power distribution results corresponding to each spatial analysis window are correspondingly organized with the coordinate position data of each domain, the regional range data, and the terrain raster distribution information to generate frequency band response data.

[0041] It should also be noted that the wavelength intervals are divided according to integer multiples of the grid resolution of each domain, and the boundary between two adjacent wavelength intervals is determined by the minimum and maximum wavelengths that can be distinguished at the corresponding analysis scale, so as to ensure that the topographic power spectrum analysis results under different domain layers are comparable.

[0042] S2.3. Based on the nested hierarchical relationship between parent and child domains, perform position registration and scale comparison of the frequency band response data between different domain layers to extract the terrain power spectrum features at different spatial scales.

[0043] Specifically, based on the nested hierarchical relationship between parent and child domains, the coordinate position data and terrain raster distribution information corresponding to each domain in the frequency band response data are hierarchically matched to determine the frequency band response results corresponding to the parent and child domains at the same spatial location. According to the grid resolution of each domain and the analysis scale corresponding to each layer of spatial analysis window, the frequency band response results that have completed hierarchical matching are scaled and compared. Based on the power distribution location, power concentration degree and dominant undulation interval within the corresponding wavelength interval, the wavelength distribution and power distribution corresponding to different domain layers at the same spatial location are obtained. The frequency band response results that have completed location registration and scale comparison are merged layer by layer according to spatial scale to extract the terrain power spectrum features at different spatial scales.

[0044] S2.4. Perform scale classification and hierarchical correspondence on the topographic power spectrum characteristics at different spatial scales to generate hierarchical frequency band identification data.

[0045] Specifically, the topographic power spectrum features at different spatial scales are grouped according to their corresponding spatial scale, wavelength distribution, and power distribution to form frequency band correspondence results at the same scale. Based on the nested hierarchical relationship between parent and child domains in the model grid parameters and the grid resolution of each domain, the frequency band correspondence results at the same scale are mapped to the domain level to form hierarchical frequency band correspondence results. The hierarchical frequency band correspondence results are associated and marked with coordinate location data and topographic raster distribution information in spatial scale order to clarify the scale and hierarchical position of the topographic power spectrum features at different spatial scales in each domain level, and generate hierarchical frequency band identification data.

[0046] S2.5. Based on the resolution level of the model grid parameters, configure the corresponding topographic spectrum retention requirements for different wavelength intervals in the layered frequency band identification data, and generate spectrum constraint data.

[0047] Specifically, based on the resolution hierarchy of the model grid parameters, the wavelength intervals corresponding to each domain in the layered frequency band identification data are hierarchically sorted to clarify the corresponding position of each wavelength interval in different domain layers; according to the matching relationship between the grid resolution of each domain and the corresponding wavelength interval, the topographic spectral retention requirement of each domain for the corresponding wavelength interval is calculated, and the expression is as follows: ; in, Indicates the first The domain is the first Topographic spectral retention requirements for each wavelength range Indicates the domain layer number, Indicates the wavelength range number. Indicates the first The domain in the first Frequency response values ​​for each wavelength range Indicates the first The sum of the frequency band response values ​​of each domain across the entire wavelength range. Indicates wavelength range index. Indicates the first Representative wavelengths for each wavelength range Indicates the resolution matching coefficient. Indicates the first The grid resolution of each domain. This represents a small constant used to prevent the denominator from being zero. This represents the natural exponential function; where the resolution matching coefficient is determined based on the ratio of the target domain grid resolution to the wavelength representing the corresponding wavelength range. Based on the topographic spectrum preservation requirements, preservation constraint marks are assigned to different wavelength intervals in the layered frequency band identification data, so that different wavelength intervals correspond to the topographic spectrum preservation requirements of each domain. The wavelength intervals with completed preservation constraint marks are associated and organized with the spatial scale information, coordinate location data and topographic grid distribution information in the layered frequency band identification data to generate spectrum constraint data, which includes wavelength interval location, layer affiliation and topographic spectrum preservation requirements.

[0048] S2.6. Based on the terrain location of each domain in the 100-meter-level basic terrain set, perform domain-specific correspondence and encapsulation of the spectral constraint data to generate a terrain spectral constraint terrain set.

[0049] Specifically, based on the coordinate location data, regional range data, and terrain grid distribution information corresponding to each domain in the 100-meter-level basic terrain set, the wavelength interval positions and hierarchical assignments in the spectral constraint data are divided into domains for location, so that the spectral constraint data corresponds to the terrain positions of each domain. According to the spatial distribution relationship corresponding to the terrain positions of each domain, the spectral constraint data that has been divided into domains is classified and organized to ensure that the spectral constraint data in each domain is consistent with the corresponding elevation data and coordinate location data. The spectral constraint data that has been divided into domains is encapsulated according to the boundaries of each domain to generate a terrain spectral constraint terrain set, which includes the terrain position of each domain, wavelength interval position, hierarchical assignment, and terrain spectral retention requirements.

[0050] S3. Based on the terrain set constrained by the terrain wave spectrum, analyze the changes in terrain power spectrum under the action of terrain smoothing function, and adjust the terrain smoothing function to generate improved smoothing function configuration data.

[0051] S3.1 The terrain smoothing function is the terrain smoothing function in the preprocessing program of the numerical weather prediction model.

[0052] Specifically, in the numerical weather prediction model preprocessing program, according to the nested hierarchical relationship between parent and child domains in the model grid parameters, the grid resolution, grid range, and projection parameters of each domain, the corresponding neighborhood relationship of the grids involved in terrain preprocessing is defined, and the smoothing calculation relationship between the central grid point and adjacent grid points is uniformly configured. This ensures that the terrain smoothing function corresponds to the common position and range of action of the target area's 100-meter-level original terrain data, 100-meter-level basic terrain set, and improved smoothing function configuration data. This clarifies the smoothing calculation basis in the terrain preprocessing stage, resulting in the terrain smoothing function in the numerical weather prediction model preprocessing program.

[0053] S3.2 Based on the nested hierarchical relationship between the parent domain and the child domain and the correspondence between the pattern regions, extract the spectral constraint information from the topographic spectral constraint terrain set, and perform domain-specific correspondence and consolidation to generate regional spectral constraint data.

[0054] Specifically, based on the nested hierarchical relationship between parent and child domains, the hierarchical affiliation of each domain in the topographic spectral constraint terrain set is screened layer by layer. According to the model region correspondence, the wavelength interval position, topographic spectral retention requirements, and coordinate position data within each domain are located to obtain the spectral constraint information corresponding to each domain. According to the hierarchical order of parent and child domains, the spectral constraint information corresponding to each domain is matched with the boundary position of each domain, so that the wavelength interval position, hierarchical affiliation, and topographic spectral retention requirements fall into the corresponding domain layer. The spectral constraint information that has been matched with the domain layer is consolidated by domain layer, and the consistency between the wavelength interval position, hierarchical affiliation, topographic spectral retention requirements, and model region correspondence is maintained to generate regional spectral constraint data.

[0055] S3.3. Based on the regional spectral constraint data, perform intensity and wavelength distribution comparisons on the topographic power spectrum under the action of the topographic smoothing function to generate regional power spectrum comparison data.

[0056] Specifically, based on the wavelength interval locations, topographic spectrum preservation requirements, and hierarchical classification in the regional spectral constraint data, a trial smoothing function is applied to the corresponding elevation data of the 100-meter-level basic topographic set to generate trial smoothed topographic data. The elevation values ​​of each domain in the trial smoothed topographic data are then segmented and mean-removed according to spatial analysis windows. A two-dimensional discrete Fourier transform is performed on the two-dimensional elevation distribution within each spatial analysis window. Based on the transform results, the power values ​​corresponding to each wavenumber component are calculated. Then, combined with the grid resolution of each domain, each wavenumber component is converted into its corresponding wavelength interval. The power values ​​within the same wavelength interval are then accumulated segment by segment to obtain the final smoothed topographic data. The trial power spectrum results for each domain corresponding to the smoothing function are obtained. Within the power spectrum comparison range corresponding to each domain, the trial power spectrum results are compared with the target retention requirements corresponding to the regional spectral constraint data according to the wavelength interval position, and the power distribution intensity difference results in each wavelength interval are obtained to form intensity comparison results. According to the hierarchical classification and wavelength interval position, the power peak position, power concentration interval and wavelength expansion position in each wavelength interval are compared to form wavelength distribution comparison results. The intensity comparison results and wavelength distribution comparison results are linked and organized according to the domain boundary, hierarchical classification and wavelength interval position to generate regional power spectrum comparison data.

[0057] It should also be noted that the trial smoothed terrain data is only used to analyze the impact of the terrain smoothing function on the terrain power spectrum, and is not used as the terrain input result for the final writing mode preprocessing interface.

[0058] S3.4. Based on the regional power spectrum comparison data, identify the differences in wavelength distribution and intensity distribution of the topographic power spectrum under the action of the topographic smoothing function, and generate regional power spectrum difference data.

[0059] Specifically, the intensity comparison results and wavelength distribution comparison results are arranged by domain according to the domain boundaries, hierarchical classification, and wavelength interval positions in the regional power spectrum comparison data. Within each wavelength interval position, the changes in power distribution intensity in the intensity comparison results and the changes in power concentration positions in the wavelength distribution comparison results are compared to obtain the difference positions and degrees of difference for each domain in each wavelength interval position. The difference positions and degrees of difference are categorized and organized according to hierarchical classification and wavelength interval positions, and are marked accordingly with the topographic wave spectrum preservation requirements in the regional wave spectrum constraint data. The difference positions, degrees of difference, hierarchical classification, and wavelength interval positions that have been marked are associated and organized according to the domain boundaries to generate regional power spectrum difference data.

[0060] S3.5. Difference segmentation is performed on the regional power spectrum difference data to generate smooth difference segmentation data.

[0061] Specifically, based on the domain boundaries, hierarchical affiliation, wavelength interval location, and degree of difference in the regional power spectrum difference data, spatially continuous difference locations with similar degrees of difference are merged to form difference segments. According to the wavelength interval location and hierarchical affiliation corresponding to each difference segment, the starting position, ending position, and extension range of each difference segment within each domain are segmented and marked, forming segmentation marking results. These segmentation marking results are then correlated with the topographic wave spectrum preservation requirements, difference locations, and degree of difference in the regional power spectrum difference data, ensuring that each segment simultaneously possesses domain layer location, wavelength interval location, and degree of difference identification, generating smooth difference segmented data.

[0062] It should also be noted that similar degree of difference means that the difference in absolute value of the degree of difference between adjacent difference positions does not exceed the preset difference segment threshold and is in the same difference level range, and is judged to be similar degree of difference; wherein, the preset difference segment threshold is set according to the statistical distribution of the power deviation rate of each domain and the sensitivity of difference fluctuation within the wavelength range, for example, preferably 0.05 to 0.15.

[0063] S3.6. Based on the smoothing difference segmented data, adjust the weight configuration of the central grid point and adjacent grid points within the terrain smoothing function to generate smoothing function weight data.

[0064] Specifically, based on the domain boundaries, hierarchical affiliation, wavelength interval location, degree of difference, and segmentation labeling results in the smoothed difference segmented data, the application location of the terrain smoothing function is located segment by segment, clarifying the range of the central grid point and adjacent grid points corresponding to each segment, and generating segmented weight positioning results. According to the degree of difference and terrain spectrum preservation requirements corresponding to each segment in the segmented weight positioning results, the existing weight configuration of the central grid point and adjacent grid points within the terrain smoothing function is adjusted segment by segment, so that the weight of the central grid point and the weight of the adjacent grid point form a matching relationship within the corresponding segment, generating segmented weight adjustment results. The segmented weight adjustment results are organized according to the domain boundaries, hierarchical affiliation, and wavelength interval location, and are kept consistent with the segmentation labeling results in the smoothed difference segmented data to generate smoothing function weight data.

[0065] It should also be noted that when the power attenuation degree of the corresponding wavelength interval in the segmented weight positioning result is higher than the preset power attenuation threshold, the weight of the center grid point in the corresponding segment is increased and the weight of adjacent grid points is decreased; when the power attenuation degree of the corresponding wavelength interval is lower than or equal to the preset power attenuation threshold, the existing weight configuration of the center grid point and adjacent grid points in the terrain smoothing function remains unchanged; and the sum of the adjusted center grid point weight and the weights of each adjacent grid point remains 1. The preset power attenuation threshold is used to characterize the degree of power attenuation of the corresponding wavelength interval power value after the terrain smoothing function is applied relative to the power value before smoothing, and is preferably set to 0.10 to 0.20.

[0066] S3.7. Based on the terrain set constrained by the terrain spectrum, perform domain-specific correspondence and integration of the smoothing function weight data to generate improved smoothing function configuration data.

[0067] Specifically, based on the topographic location, wavelength range location, hierarchical affiliation, and topographic spectrum preservation requirements of each domain in the topographic spectrum-constrained topographic set, the weights of the central grid points and adjacent grid points in the smoothing function weight data are located and matched domain by domain to generate domain-specific weight correspondence results. According to the domain-specific weight correspondence results, the weights of the central grid points and adjacent grid points in each domain are classified and merged according to wavelength range location and hierarchical affiliation, so that the smoothing function weight data and the topographic spectrum preservation requirements of the topographic spectrum-constrained topographic set form a consistent correspondence, generating domain-specific weight integration results. Based on the domain-specific weight integration results, the weights of the central grid points, adjacent grid points, wavelength range location, hierarchical affiliation, and topographic spectrum preservation requirements in each domain are jointly organized to generate improved smoothing function configuration data.

[0068] S4. Based on the improved smoothing function configuration data, smooth the 100-meter-level basic terrain set to generate spectral-fidelity smoothed terrain data.

[0069] S4.1. Based on the improved smoothing function configuration data, perform a smoothing transformation on the elevation data of the 100-meter-level basic terrain set to generate initial smoothed terrain data.

[0070] Specifically, based on the weights of the central grid points, adjacent grid points, wavelength intervals, and hierarchical affiliations in the improved smoothing function configuration data, the elevation data of the 100-meter-level basic terrain set is segmented and located, and each terrain grid point is mapped to the corresponding weight position to generate weight-corresponding elevation segments. According to the adjacency relationship between each terrain grid point in the weight-corresponding elevation segment, the elevations of the central grid point and adjacent grid points are weighted and smoothed to generate segmented smoothed elevation results. The segmented smoothed elevation results are then written back to the corresponding terrain grid points according to the coordinate position data, regional range data, and terrain grid distribution information of the 100-meter-level basic terrain set to generate initial smoothed terrain data.

[0071] S4.2. Based on the nested hierarchical relationship between parent and child domains and the grid resolution of each domain in the model grid parameters, the initial smoothed terrain data is divided into domains and aligned with their positions to generate domain-smoothed terrain data.

[0072] Specifically, based on the nested hierarchical relationship between parent and child domains in the model grid parameters, the coordinate position data, regional range data, and terrain raster distribution information in the initial smoothed terrain data are divided into domain layers to form domain layer attribution results. According to the grid resolution of each domain, the elevation data in the domain layer attribution results are merged according to the corresponding grid scale, and the merged elevation data is matched point by point with the corresponding coordinate position data to form grid position correspondence results. The elevation data, coordinate position data, regional range data, and terrain raster distribution information in the grid position correspondence results are then sorted into domains according to the nested hierarchical relationship between parent and child domains to generate domain-specific smoothed terrain data.

[0073] S4.3. Based on the terrain set constrained by the terrain spectrum, the domain-smoothed terrain data is integrated to generate spectral-fidelity smoothed terrain data.

[0074] Specifically, based on the topographic location, wavelength interval location, hierarchical affiliation, and topographic spectrum preservation requirements of each domain in the topographic set constrained by the topographic spectrum, the elevation data in the domain-smoothed topographic data is located and compared domain by domain to generate domain-specific spectrum correspondence results. According to the domain-specific spectrum correspondence results, the distribution of elevation data in each domain at the corresponding wavelength interval location is merged. The merged elevation data is then consistent with the corresponding coordinate location data, regional range data, and topographic raster distribution information to generate domain-specific spectrum integration results. When the power distribution of the corresponding wavelength interval does not meet the topographic spectrum preservation requirements, a secondary weight correction or local resmoothing is performed on the elevation data of the corresponding domain. The domain-specific spectrum integration results are then jointly organized according to the domain boundaries and hierarchical affiliation to generate spectrum-fidelity smoothed topographic data.

[0075] S5. Perform terrain power spectrum consistency check and terrain height deviation check on the spectral fidelity smoothed terrain data to generate a spectral fidelity model terrain field.

[0076] S5.1. Perform domain-by-domain topographic power spectrum comparison between the spectral fidelity smoothed topographic data and the 100-meter-level basic topographic set to generate domain-by-domain spectral comparison data.

[0077] Specifically, based on the nested hierarchical relationship between parent and child domains in the model grid parameters, the coordinate position data, regional range data, and terrain raster distribution information of the spectral fidelity smoothed terrain data and the 100-meter-level basic terrain set are registered separately to form a domain alignment result. According to the domain alignment result, the elevation data corresponding to the spectral fidelity smoothed terrain data and the elevation data corresponding to the 100-meter-level basic terrain set are extracted in each domain, and the terrain power spectrum is calculated according to the same spatial location and the same analysis scale to form a domain power spectrum calculation result. Based on the domain power spectrum calculation result, according to the boundary and wavelength interval of each domain, the power distribution corresponding to the spectral fidelity smoothed terrain data and the power distribution corresponding to the 100-meter-level basic terrain set are arranged zone by zone. The power distribution strength, power peak position, power concentration interval, and wavelength expansion position within the same wavelength interval are compared one by one, and the power offset relationship and concentration degree change of each domain at different wavelength interval positions are recorded to form a domain spectral comparison result. The domain spectral comparison result is organized according to the boundary, hierarchical affiliation, and wavelength interval position of each domain to generate domain-by-domain spectral comparison data.

[0078] S5.2. Based on the topographic spectrum constraining the topographic set, perform consistency verification on the domain-by-domain spectrum comparison data to generate spectrum consistency data.

[0079] Specifically, based on the topographic location, wavelength interval location, hierarchical affiliation, and topographic spectrum preservation requirements of each domain in the topographic spectrum constraint topographic set, the domain-specific spectrum comparison results in the domain-by-domain spectrum comparison data are located domain by domain to form spectrum constraint correspondence results. According to the spectrum constraint correspondence results, the power distribution intensity, power peak location, power concentration interval, and wavelength expansion location corresponding to each wavelength interval in the domain-specific spectrum comparison results are checked against the topographic spectrum preservation requirements item by item, and the power deviation of the corresponding wavelength interval is calculated. When the power deviation of the corresponding wavelength interval does not exceed the preset spectrum deviation threshold, the consistency status of the corresponding wavelength interval is determined to be consistent; otherwise, the consistency status of the corresponding wavelength interval is determined to be inconsistent, forming a spectrum consistency verification result. The domain boundaries, hierarchical affiliation, wavelength interval location, and consistency status in the spectrum consistency verification result are correspondingly organized and kept consistent with the domain-specific spectrum comparison results in the domain-by-domain spectrum comparison data to generate spectrum consistency data.

[0080] It should also be noted that the spectral deviation threshold is set based on the statistical distribution of the power deviation rate within the corresponding wavelength range of different domain layers and the sensitivity of topographic spectral retention, for example, preferably 0.10 to 0.20.

[0081] S5.3. Based on the spectral consistency data and the elevation data of the 100-meter-level basic terrain set, the terrain height deviation of the spectral fidelity smoothed terrain data is checked to generate spectral checked terrain data.

[0082] Specifically, based on the domain boundaries, hierarchical affiliation, wavelength interval location, and consistency status in the spectral consistency data, the elevation data of the corresponding spatial location in the spectral fidelity smoothed terrain data are located domain by domain to generate elevation segments to be checked. According to the spatial location corresponding to the elevation segment to be checked, the corresponding elevation value is extracted from the elevation data of the 100-meter-level basic terrain set, and the deviation between the corresponding elevation value and the elevation value in the elevation segment to be checked is calculated point by point to generate terrain height deviation results. When the elevation deviation of the corresponding spatial location exceeds the preset height deviation threshold, the elevation value of the corresponding location is corrected or marked as a location to be resmoothed. Based on the consistency status in the spectral consistency data, the terrain height deviation results are classified and organized according to the domain boundaries, hierarchical affiliation, and wavelength interval location, and the classified and organized elevation values, deviation values, locations to be resmoothed, and corresponding spatial locations are jointly organized to generate spectral verification terrain data.

[0083] It should also be noted that the height deviation threshold is set based on the acceptable range of terrain elevation error corresponding to the grid resolution of each domain and the fluctuation range of the original terrain elevation. For example, it is preferable to take 0.25 to 0.50 times the corresponding domain grid resolution.

[0084] S5.4 Perform cross-domain consistency verification and merging on the spectral verification terrain data to generate a spectral fidelity model terrain field.

[0085] Specifically, based on the domain boundaries, hierarchical affiliation, wavelength interval positions, elevation values, and deviation values ​​in the spectral verification terrain data, the elevation values ​​and deviation values ​​of corresponding spatial positions in the overlapping areas of the parent and child domains are compared point by point to form cross-domain comparison results. According to the cross-domain comparison results, the continuity of elevation values, deviation values, and wavelength interval positions in the same spatial positions of the parent and child domains are checked for consistency to form cross-domain consistency verification results. If the cross-domain consistency verification results fail, the process is returned to readjust the smoothing function weight configuration, or the corresponding subdomain is re-smoothed and the corresponding spatial positions after reprocessing are checked for cross-domain consistency again. Based on the cross-domain consistency verification results, the elevation values, coordinate position data, regional range data, and terrain raster distribution information that have passed the consistency verification are merged and organized in domain layer order, while maintaining the correspondence between the domain boundaries, hierarchical affiliation, and wavelength interval positions, to generate a spectral fidelity model terrain field.

[0086] S6. Write the spectral fidelity model topographic field into the numerical weather prediction model preprocessing interface to generate a spectral fidelity preprocessed topographic set.

[0087] S6.1 The numerical weather prediction model preprocessing interface is generated by the numerical weather prediction model preprocessing program based on the nested hierarchy of parent and child domains, the grid resolution of each domain, the grid range, and the projection parameters in the model grid parameters.

[0088] Specifically, the preprocessing program of the numerical weather prediction model reads the nested hierarchical relationship between parent and child domains in the model grid parameters to clarify the hierarchical organization order between each domain; according to the grid resolution, grid range, and projection parameters of each domain, the corresponding grid coverage area, spatial projection representation, and data writing position of each domain are configured accordingly to form a preprocessing input relationship consistent with the hierarchical position of each domain; the configured parent and child domain hierarchical information, grid position information, and projection position information are organized according to the data access format of the preprocessing program to generate the numerical weather prediction model preprocessing interface.

[0089] S6.2. Based on the numerical weather prediction model preprocessing interface, perform grid adaptation and structural reorganization on the topographic field of the spectral fidelity model to generate preprocessed adapted topographic data.

[0090] Specifically, based on the grid access order, grid coverage location, and projection representation method of each domain in the numerical weather prediction model preprocessing interface, the elevation values, coordinate location data, regional range data, and terrain raster distribution information in the spectral fidelity model terrain field are read and their locations are verified domain by domain to form the terrain sequence corresponding to the interface. According to the hierarchical arrangement relationship between the parent and child domains corresponding to the numerical weather prediction model preprocessing interface, the data of each domain in the terrain sequence corresponding to the interface are rearranged hierarchically and the boundaries are connected and organized to form a hierarchically reconstructed terrain result. Based on the grid cell organization format corresponding to the numerical weather prediction model preprocessing interface, the elevation values, coordinate location data, regional range data, and terrain raster distribution information in the hierarchically reconstructed terrain result are written into a data structure consistent with the preprocessing access format to generate preprocessing adapted terrain data.

[0091] S6.3. Based on the nested hierarchical relationship between parent and child domains, the terrain set constrained by the terrain spectrum, and the configuration data of the improved smoothing function, perform a consistency check on the pre-processed adaptive terrain data before writing, and generate pre-processed verification terrain data.

[0092] Specifically, based on the nested hierarchical relationship between parent and child domains, the arrangement order, hierarchical position, and boundary connection position of each domain in the preprocessed adapted terrain data are checked domain by domain to generate hierarchical verification results. Based on the terrain spectrum constraint terrain set, the wavelength interval position and terrain spectrum retention requirements corresponding to each domain in the hierarchical verification results are checked item by item. Based on the improved smoothing function configuration data, the elevation value configuration relationship of each corresponding position in the preprocessed adapted terrain data is checked segment by segment to generate constraint verification results. If the consistency check before writing fails, the process is returned to re-adapt the grid and restructure the data, and the hierarchical and constraint checks are performed again on the re-processed preprocessed adapted terrain data. The hierarchical verification results and constraint verification results are then organized according to the domain boundaries, hierarchical affiliation, wavelength interval position, and elevation value position to generate preprocessed verification terrain data.

[0093] S6.4. Based on the numerical weather prediction model preprocessing interface, the preprocessed verification terrain data is organized into domains and written into the numerical weather prediction model preprocessing interface to generate a spectral fidelity preprocessed terrain set.

[0094] Specifically, based on the access order, grid coverage location, and projection correspondence of each domain in the numerical weather prediction model preprocessing interface, the elevation values, coordinate location data, regional range data, and terrain raster distribution information in the preprocessed verification terrain data are split and arranged sequentially domain by domain to generate a domain-based organization result. According to the domain layer arrangement structure and data access format corresponding to the numerical weather prediction model preprocessing interface, the domain data in the domain-based organization result are matched for writing location and boundary alignment to generate an interface writing result. The elevation values, coordinate location data, regional range data, and terrain raster distribution information in the interface writing result are written into the numerical weather prediction model preprocessing interface in domain layer order to generate a spectral fidelity preprocessed terrain set.

[0095] In summary, this invention achieves differentiated adjustment of the terrain power spectrum under the action of the terrain smoothing function by analyzing the terrain power spectrum change based on the terrain set constrained by the terrain spectrum and generating improved smoothing function configuration data. This enables the terrain smoothing function to be used to adjust the terrain power spectrum for different terrain spectral components in a differentiated manner, so that the smoothing process and the spectral preservation requirements corresponding to the model grid parameters form a consistent constraint. This data is used to generate spectrally accurate smoothed terrain data and support the construction of spectrally accurate model terrain fields. This results in improved model terrain fineness, spectral consistency, and adaptability of numerical weather prediction model preprocessing.

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for preserving refined terrain features at the 100-meter level, characterized in that: include, Collect raw terrain data and model grid parameters at the 100-meter level, and perform unified projection transformation and grid mapping processing to generate a basic terrain set at the 100-meter level; Topographic power spectrum features at different spatial scales are extracted from the basic topographic set at the hundred-meter level, and the corresponding topographic wave spectrum preservation requirements are configured according to the resolution level of the model grid parameters to generate a topographic wave spectrum constrained topographic set. Based on the terrain set constrained by the terrain wave spectrum, the changes in terrain power spectrum under the action of terrain smoothing function are analyzed, and the terrain smoothing function is adjusted to generate improved smoothing function configuration data. Based on the improved smoothing function configuration data, the 100-meter-level basic terrain set is smoothed to generate spectral-fidelity smoothed terrain data. Perform terrain power spectrum consistency verification and terrain height deviation verification on the spectral fidelity smoothed terrain data to generate a spectral fidelity model terrain field; Write the spectral fidelity model topographic field into the numerical weather prediction model preprocessing interface to generate a spectral fidelity preprocessed topographic set.

2. The method for preserving refined terrain features at the 100-meter level as described in claim 1, characterized in that: The raw terrain data at the hundred-meter level includes elevation data, coordinate location data, regional range data, and terrain grid distribution information; The pattern grid parameters include the nesting hierarchy between parent and child domains, the grid resolution of each domain, the grid range, the grid number, the projection parameters, and the corresponding relationship of the pattern regions; The terrain smoothing function is the terrain smoothing function in the numerical weather prediction model preprocessing program. The numerical weather prediction model preprocessing interface is generated by the numerical weather prediction model preprocessing program based on the nested hierarchy of parent and child domains, the grid resolution of each domain, the grid range, and the projection parameters in the model grid parameters.

3. The method for preserving refined terrain features at the 100-meter level as described in claim 2, characterized in that: The generation of the 100-meter-level basic terrain dataset specifically includes, The raw terrain data at the 100-meter level is transformed using a unified projection datum to generate unified projection terrain data. Based on the grid range, grid number and pattern region correspondence in the model grid parameters, the unified projection terrain data is divided into domains, labeled with regional affiliation and raster position correspondence, and generated into domain-mapped terrain data. Based on the nested hierarchical relationship between parent and child domains in the model grid parameters, the grid resolution and projection parameters of each domain, cross-domain position alignment, resolution adaptation and hierarchical association are performed on the domain-mapped terrain data to generate a 100-meter-level basic terrain set.

4. The method for preserving refined terrain features at the 100-meter level as described in claim 1, characterized in that: The extraction of terrain power spectrum features at different spatial scales specifically includes... The topographic elevation distribution of each domain is extracted from the basic topographic data at the 100-meter level, and the analysis scale is set according to the grid resolution of each domain at different spatial scales to generate multi-scale topographic data of different domains. Scale separation and power spectrum analysis of terrain elevation undulation within each spatial analysis window of multi-scale terrain data in the domain are performed to generate frequency band response data. Based on the nested hierarchical relationship between parent and child domains, the frequency band response data is registered and scaled between different domain layers to extract the topographic power spectrum features at different spatial scales.

5. The method for preserving refined terrain features at the 100-meter level as described in claim 4, characterized in that: The generated topographic spectrum-constrained terrain set specifically includes... The topographic power spectrum features at different spatial scales are classified by scale and hierarchically mapped to generate hierarchical frequency band identification data. Based on the resolution level of the model grid parameters, the corresponding topographic spectrum retention requirements are configured for different wavelength intervals in the layered frequency band identification data to generate spectrum constraint data. Based on the terrain location of each domain in the 100-meter-level basic terrain set, the spectral constraint data is divided into domains and encapsulated to generate a terrain spectral constraint terrain set.

6. The method for preserving refined terrain features at the 100-meter level as described in claim 1, characterized in that: The analysis of the terrain power spectrum changes under the action of the terrain smoothing function specifically includes: Based on the nested hierarchical relationship between parent and child domains and the correspondence between pattern regions, spectral constraint information is extracted from the topographic spectral constraint terrain set, and then domain-specific correspondence and consolidation are performed to generate regional spectral constraint data. Based on the regional spectral constraint data, intensity and wavelength distribution comparisons are performed on the topographic power spectrum under the action of the topographic smoothing function to generate regional power spectrum comparison data. Based on regional power spectrum comparison data, the differences in wavelength and intensity distribution of the topographic power spectrum under the action of the topographic smoothing function are identified, and regional power spectrum difference data are generated.

7. The method for preserving refined terrain features at the 100-meter level as described in claim 6, characterized in that: The configuration data for generating the improved smoothing function specifically includes, The regional power spectrum difference data is segmented and labeled to generate smooth difference segment data; Based on the smoothing difference segmented data, the weight configuration of the central grid point and adjacent grid points within the terrain smoothing function is adjusted to generate smoothing function weight data; Based on the terrain set constrained by the topographic wave spectrum, the smoothing function weight data is divided into domains and integrated to generate improved smoothing function configuration data.

8. The method for preserving refined terrain features at the 100-meter level as described in claim 7, characterized in that: The generation of spectrally fidelity smoothed terrain data specifically includes, Based on the improved smoothing function configuration data, the elevation data of the 100-meter-level basic terrain set is smoothed to generate initial smoothed terrain data. Based on the nested hierarchical relationship between parent and child domains and the grid resolution of each domain in the pattern grid parameters, the initial smoothed terrain data is divided into domains and aligned with the location to generate domain smoothed terrain data. Based on the topographic spectrum constrained topographic set, the domain-smoothed topographic data is integrated to generate spectrally faithful smoothed topographic data.

9. The method for preserving refined terrain features at the 100-meter level as described in claim 8, characterized in that: The generated spectral fidelity model topographic field specifically includes, A domain-by-domain topographic power spectrum comparison was performed between the spectral-fidelity smoothed topographic data and the 100-meter-level basic topographic set to generate domain-by-domain spectral comparison data. Based on the topographic spectral constraint topographic set, the consistency of the domain-by-domain spectral comparison data is verified to generate spectral consistency data. Based on the spectral consistency data and the elevation data of the 100-meter-level basic terrain set, the terrain height deviation of the spectral fidelity smoothed terrain data is checked to generate spectral checked terrain data. Cross-domain consistency verification and merging are performed on the spectral verification terrain data to generate a spectral fidelity model terrain field.

10. The method for preserving refined terrain features at the 100-meter level as described in claim 9, characterized in that: The generated spectral fidelity preprocessing terrain set specifically includes... Based on the numerical weather prediction model preprocessing interface, the topographic field of the spectral fidelity model is meshed and restructured to generate preprocessed adapted topographic data. Based on the nested hierarchical relationship between parent and child domains, the terrain set constrained by the terrain spectrum, and the configuration data of the improved smoothing function, the pre-processed adaptive terrain data is checked for consistency before being written, and pre-processed verification terrain data is generated. Based on the numerical weather prediction model preprocessing interface, the preprocessed verification terrain data is organized by domain and written into the numerical weather prediction model preprocessing interface to generate a spectral fidelity preprocessed terrain set.

Images