A method for tropical cyclone wind field simulation by assimilating observations and spectral relaxation

Abstract

The application relates to a tropical cyclone wind field simulation method fusing observation and spectral relaxation assimilation. The application is suitable for the fields of meteorological numerical simulation and new energy engineering technology. The technical scheme comprises the following steps: determining a simulation region according to the path and influence range of a target tropical cyclone, and obtaining multi-source observation data during the target tropical cyclone; constructing a first numerical simulation experiment by using a mesoscale numerical model, assimilating the multi-source observation data into the mesoscale numerical model by using an observation approximation assimilation method, and obtaining a first wind field simulation result; constructing a second numerical simulation experiment on the basis of the first numerical simulation experiment, determining spectral relaxation parameter configuration based on the first wind field simulation result by using a spectral relaxation assimilation method, relaxing a large-scale circulation field in the model to large-scale reanalysis driving data, and obtaining a second wind field simulation result; and generating tropical cyclone wind field simulation data based on the second wind field simulation result.

Description

Technical Field

[0001] This invention relates to a method for simulating tropical cyclone wind fields by integrating observation and spectral relaxation assimilation. It is applicable to the fields of meteorological numerical simulation and new energy engineering technology. Background Technology

[0002] Tropical cyclones are one of the meteorological hazards affecting offshore wind power development, directly impacting wind farm site selection, turbine selection, structural design, and long-term operation and maintenance. Accurate assessment of tropical cyclone wind fields is fundamental to engineering decisions; however, observational data during tropical cyclones is severely lacking, making it difficult to obtain high-wind samples that meet engineering reliability requirements.

[0003] To compensate for the inadequacy of observational data, numerical simulation has become an important alternative data source. Currently, the global reanalysis data (such as ERA5) widely used in the wind energy field systematically underestimates the intensity of tropical cyclones, making it difficult to meet the accuracy requirements involved in engineering projects. Mesoscale meteorological models (such as WRF) offer higher resolution and more comprehensive parameterization of physical processes, demonstrating advantages in simulating tropical cyclones.

[0004] However, achieving high-precision tropical cyclone simulations using wind-radio frequency (WRF) remains challenging. The accuracy of tropical cyclone simulations is influenced by both their inherent complex dynamic processes and their interactions with the large-scale environment. These two factors often make it difficult for models to simultaneously guarantee the accuracy of both track and intensity simulations. Furthermore, existing methods primarily focus on wind speeds at a height of 10 m, lacking direct simulation and validation of strong wind processes at turbine hub height (50–200 m), and even more so, lacking accurate wind measurement data interpolation methods for tropical cyclone formation, and a systematic solution for quantitative assessment of strong wind disaster risks.

[0005] As my country's offshore wind power continues to expand into deeper and more distant waters, the impact of tropical cyclones is becoming increasingly prominent. Therefore, there is an urgent need to develop a highly reliable numerical simulation method that can balance the accuracy of tropical cyclone path and intensity simulations, particularly optimizing the simulation of high-wind processes at hub height, and directly outputting the wind parameter products required for engineering projects. This will support the scientific planning, safe design, and robust operation of offshore wind power projects. Summary of the Invention

[0006] The technical problem to be solved by this invention is to provide a method for simulating tropical cyclone wind fields by integrating observation and spectral relaxation assimilation, in view of the above-mentioned problems.

[0007] The technical solution adopted in this invention is: a method for simulating tropical cyclone wind fields by integrating observation and spectral relaxation assimilation, comprising: Based on the path and impact range of the target tropical cyclone, the simulation area is determined, and multi-source observation data during the target tropical cyclone are obtained. This multi-source observation data includes at least large-scale reanalysis driving data and wind field observation data within the wind turbine hub height range. The first numerical simulation experiment was constructed using a mesoscale numerical model. The observation approximation assimilation method was used to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results. Based on the first numerical simulation experiment, a second numerical simulation experiment is constructed. The spectral relaxation assimilation method is adopted. Based on the first wind field simulation results, the spectral relaxation parameter configuration is determined. The large-scale circulation field in the model is driven to the large-scale reanalysis to relax the data, and the second wind field simulation results are obtained. Based on the second wind field simulation results, tropical cyclone wind field simulation data are generated; The mesoscale numerical model contains at least horizontally nested parent and child domain computational grids in the horizontal direction; the spectral relaxation assimilation is only enabled in the parent domain.

[0008] The multi-source observation data also includes large-scale integrated meteorological observation data and / or high-resolution sea surface temperature data.

[0009] The method of assimilating multi-source observation data into a mesoscale numerical model using observation approximation includes: Based on the configured mode, a set of observational approximation assimilation experiments is constructed. This set of experiments includes at least a control group and multiple assimilation experiments, which are constructed by adjusting the combination of assimilation data and assimilation parameters. After running this set of experiments, the simulation accuracy of path error, minimum central pressure, and maximum near-center wind speed of each experimental result was comprehensively evaluated based on reliable tropical cyclone optimal path and intensity observation datasets. Based on the evaluation results, the experiment that best simulated the intensity of tropical cyclones was selected from this group of experiments and used as the first wind field simulation result.

[0010] The method employing spectral relaxation assimilation, based on the first wind field simulation results, determines the spectral relaxation parameter configuration, and relaxes the large-scale circulation field of the mesoscale numerical model towards the large-scale reanalysis-driven data, including: Based on the simulation results of the first wind field, the configuration of spectral relaxation assimilation parameters for multiple sets of comparative experiments was determined, including assimilation variables and relaxation coefficients; After running this set of experiments, the improvement effect of each experiment on path error and intensity maintenance capability was evaluated based on reliable tropical cyclone optimal track and intensity observation datasets. Based on the evaluation results, the experiments that can minimize path simulation errors while maintaining or further improving intensity simulation accuracy were selected from this group of experiments and used as the second wind field simulation results.

[0011] The assimilation variables include at least the horizontal wind field component and the temperature field.

[0012] The spectral relaxation assimilation parameter configuration also includes the spectral filter cutoff wavelength, which is dynamically determined based on the parent domain range and the Rossby deformation radius of the target tropical cyclone.

[0013] A tropical cyclone wind field simulation device that integrates observation and spectral relaxation assimilation includes: The data acquisition module is used to determine the simulation area based on the path and influence range of the target tropical cyclone, and to acquire multi-source observation data during the target tropical cyclone. The multi-source observation data includes at least large-scale reanalysis driving data and wind field observation data within the wind turbine hub height range. The first assimilation simulation module is used to construct the first numerical simulation experiment using a mesoscale numerical model. It uses the observation approximation assimilation method to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results. The second assimilation simulation module is used to construct a second numerical simulation experiment based on the first numerical simulation experiment. It adopts the spectral relaxation assimilation method, determines the spectral relaxation parameter configuration based on the first wind field simulation results, and drives the data relaxation from the large-scale circulation field in the model to the large-scale reanalysis to obtain the second wind field simulation results. The simulation data generation module is used to generate tropical cyclone wind field simulation data based on the second wind field simulation results; The mesoscale numerical model contains at least horizontally nested parent and child domain computational grids in the horizontal direction; the spectral relaxation assimilation is only enabled in the parent domain.

[0014] A storage medium storing a computer program executable by a processor, wherein the computer program, when executed, implements the steps of the tropical cyclone wind field simulation method.

[0015] A tropical cyclone wind field simulation device has a memory and a processor. The memory stores a computer program that can be executed by the processor. When the computer program is executed, it implements the steps of the tropical cyclone wind field simulation method.

[0016] A method for interpolating wind measurement data in a wind farm includes: Obtain information on missing measurement periods for the wind measurement towers of the target wind farm, as well as the actual wind measurement data for adjacent non-missing measurement periods; Based on the second wind field simulation results obtained in the tropical cyclone wind field simulation method, simulated wind speed time series data corresponding to the location of the wind measurement tower are generated. Calculate the correlation coefficient between the measured wind data from the wind measurement tower during non-missing periods and the simulated wind speed time series data; When the correlation coefficient is greater than a preset threshold, the data for the missing time period is interpolated based on the simulated wind speed time series data.

[0017] A method for assessing tropical cyclone gale hazards in offshore wind power projects, comprising: Based on the second wind field simulation results obtained in the tropical cyclone wind field simulation method, spatial distribution data of the maximum wind speed of tropical cyclones at the wind turbine hub height layer are generated. The spatial distribution data of the maximum wind speed is compared with the preset wind speed threshold for the wind turbine design level; based on the comparison results, high-risk disaster areas with wind speed exceeding the standard are identified.

[0018] The beneficial effects of this invention are as follows: This invention uses multi-source observation data in conjunction with observation approximation and spectral relaxation assimilation to assimilate key data such as lidar hub height wind profile, comprehensive meteorological observation, and high-resolution sea surface temperature. Through complementary correction of multi-source data, it solves the problem of wind speed underestimation caused by the "coarse spatial resolution and insufficient parameterization of physical processes" in traditional reanalysis data, and provides highly reliable core data support for wind power projects.

[0019] This invention achieves dynamic adaptation of assimilation strategies and parameters to specific cyclone events through the process of "observation assimilation optimization of tropical cyclone intensity simulation → enabling spectral relaxation assimilation based on observation assimilation simulation results to constrain tropical cyclone path simulation", rather than using fixed empirical values, while improving the accuracy of path and intensity simulation.

[0020] This invention first optimizes the intensity of tropical cyclones by observation and approximation, then dynamically configures spectral relaxation parameters based on the intensity optimization results. By constraining large-scale circulation in the parent domain and simulating fine-scale structures in the subdomain, and by setting simulation schemes in combination with the characteristics of tropical cyclones, it accurately captures the cyclone core structure and the vertical wind profile at the hub height, thus achieving simultaneous optimization of tropical cyclone "accurate path, accurate intensity, and accurate hub height wind speed". Attached Figure Description

[0021] Figure 1 This is a flowchart of the tropical cyclone wind field simulation method in the embodiment.

[0022] Figure 2 The example uses CMA STI data to evaluate the tropical cyclone path, central minimum pressure (Pmin), and near-center maximum wind speed (Vmax) simulated by the first round of observational approximation (ON) experiment and the second round of spectral relaxation assimilation (SN) experiment.

[0023] Figure 3The following is a time series of wind speed and wind direction at 100m and 150m altitudes observed by lidar (S#, D#, T#) in the example (blue triangle symbol: lidar observation; red line: optimal assimilation scheme EXP_ON_SN_BEST; green line: control group EXP_CTRL; black "×" symbol: ERA5 reanalysis data).

[0024] Figure 4 The statistical test of the simulation results of the lidar (S#, D#, T#) at various heights from 50 to 200 m in the example is shown in the figure (red line: optimal assimilation scheme EXP_ON_SN_BEST, green line: control group EXP_CTRL, black triangle: ERA5 data).

[0025] Figure 5 This is a diagram showing the spatial distribution of maximum wind speed at a height of 100m and the risk zone division map exceeding the design wind speeds of IEC Class I (50m / s), Class II (42.5m / s), and Class III (37.5m / s) based on the optimal scheme EXP_ON_SN_BEST in the embodiment. Detailed Implementation

[0026] The embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0027] In the description of this invention, "multiple" means two or more. The use of "first" and "second" is for distinguishing technical features only and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or the order of the indicated technical features. Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0028] Example 1: As Figure 1 As shown in the figure, this embodiment is a method for simulating tropical cyclone wind fields by integrating observation and spectral relaxation assimilation, specifically including the following steps: S100. Based on the path and impact range of the target tropical cyclone, determine the simulation area and acquire multi-source observation data of the simulation area during the target tropical cyclone, including large-scale reanalysis driving data, large-scale integrated meteorological observation data, high-resolution sea surface temperature data, and wind field observation data within the wind turbine hub height range.

[0029] In this example, large-scale reanalysis driving data, such as ERA5, provides the background field and boundary conditions for the simulation; large-scale integrated meteorological observation data provides the numerical model with a comprehensive dataset covering the atmospheric state constraints of the simulation area, such as the NCEP ADP dataset which integrates multiple observation sources such as ground stations, radiosondes, ships, and buoys; high-resolution sea surface temperature data refers to data products with a spatiotemporal resolution significantly higher than the driving field that can improve the calculation of air-sea flux, such as OSTIA; and wind field observation data within the wind turbine hub height range are wind speed observation data from floating lidar and other equipment covering the wind turbine hub height range (e.g., 50~200 m).

[0030] In this embodiment, the multi-source observation data obtained above are preprocessed, including coordinate transformation, quality control, and format standardization.

[0031] S200. The first numerical simulation experiment was constructed using a mesoscale numerical model. The observation nudging (ON) method was used to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results.

[0032] In this embodiment, a mesoscale numerical model (such as WRF) is configured, and a computational grid containing at least two horizontal nested layers (such as a parent domain and a child domain with resolutions of 9 km and 3 km, respectively) is established. The vertical direction is densified near the hub height, such as setting the layer interval to 20 m below 200 m height, and a combination of physical parameterization schemes suitable for tropical cyclone simulation is selected.

[0033] S210. Based on the configured mode, construct a set of observational approximation assimilation experiments. This set of experiments includes at least a control group EXP_CTRL and multiple assimilation experiments (e.g., EXP_ON1, EXP_ON2, EXP_ON3, …).

[0034] In this embodiment, the multiple assimilation experiments are constructed by adjusting the assimilation data combination and assimilation parameters (such as weighting functions), for example, assimilating different subsets of observation data from multiple sources.

[0035] S220. After running this set of experiments, the path error, central minimum pressure (Pmin), and near-center maximum wind speed (Vmax) simulation accuracy of each experimental result are comprehensively evaluated based on reliable tropical cyclone optimal path and intensity observation datasets (such as data released by the Shanghai Typhoon Institute of China Meteorological Administration (CMA-STI) or the Joint Typhoon Warning Center (JTWC).

[0036] S230. Based on the evaluation results, the experiment that best simulates the intensity of tropical cyclones (especially Pmin and Vmax) is selected from this group of experiments and taken as the first wind field simulation result, denoted as EXP_ON_BEST.

[0037] In this embodiment, multi-source observations, especially hub height wind profile data, are used to continuously correct state variables during model integration, prioritizing the optimization of the cyclone's core intensity and vertical wind field structure.

[0038] S300. Based on the first numerical simulation experiment, a second numerical simulation experiment is constructed. The Spectral Nudging (SN) method is used to determine the spectral relaxation parameter configuration based on the first wind field simulation results. The large-scale circulation field in the model is then used to drive data relaxation through the large-scale reanalysis to obtain the second wind field simulation results.

[0039] S310. Based on the simulation results of the first wind field, determine the configuration of spectral relaxation assimilation parameters for multiple sets of comparative experiments, including assimilation variables, relaxation coefficients corresponding to the variables, and spectral filtering cutoff wavelengths, etc.

[0040] In this embodiment, the key parameter configurations for spectral relaxation assimilation in each set of comparative experiments are specifically tested and determined based on the observed and optimized wind field structure characteristics simulated by the first wind field simulation result EXP_ON_BEST. By designing multiple sets of comparative experiments (such as EXP_ON_BEST_SN1, EXP_ON_BEST_SN2, EXP_ON_BEST_SN3, …), the system tests the impact of different parameter configurations for spectral relaxation assimilation on the simulation results.

[0041] In this example, spectral relaxation assimilation is enabled only in the parent domain and disabled in nested subdomains, in order to preserve the subdomains' ability to develop fine cyclone structures while constraining large-scale circulation.

[0042] Assimilation variables and relaxation coefficients: Relaxation is applied to at least the horizontal wind field components (U, V) and the temperature field (T). Optimal ranges of relaxation coefficient values ​​are determined through comparative experiments, for example, the relaxation coefficient for the horizontal wind field. In 1×10⁻ 4 s⁻¹ to 5×10⁻ 4 Within s⁻¹, the optimal temperature field relaxation coefficient is selected. In 1×10⁻ 5 s⁻¹ to 5×10⁻ 5 s⁻¹ is the preferred option; Spectral filtering cutoff wavelength: dynamically determined based on the parent domain range and the Rossby deformation radius of the target cyclone, usually selected between 700 km and 2000 km to ensure that it mainly acts on the large-scale background field.

[0043] S320. After running this set of experiments, based on reliable datasets of optimal tropical cyclone paths and intensity observations, evaluate the improvement effect of each experiment on path error and intensity maintenance capability.

[0044] S330. Based on the evaluation results, the experiment that can minimize the path simulation error and maintain or further improve the intensity simulation accuracy is selected from this group of experiments and used as the second wind field simulation result, denoted as EXP_ON_SN_BEST.

[0045] This embodiment utilizes an optimized spectral relaxation technique to stably constrain the large-scale circulation of the mode to a more reliable driving field, thereby correcting the path deviation that may be introduced by local assimilation.

[0046] S400. Based on the second wind field simulation result EXP_ON_SN_BEST, generate tropical cyclone wind field simulation data.

[0047] In this embodiment, observation data independent of the assimilation data source (such as lidar observations that were not assimilated) is used to verify the simulation accuracy of the final scheme EXP_ON_SN_BEST. Statistical measures such as mean absolute error, bias, root mean square error, and correlation coefficient are used for quantitative evaluation to confirm that it meets the accuracy requirements for engineering applications.

[0048] In this example, for EXP_ON_SN_BEST that meets the accuracy requirements, the simulation results of EXP_ON_SN_BEST in the subdomain are extracted, and simulation data such as wind speed sequences that vary over time at a specified geographical location, and spatial distribution maps of the maximum wind speed during the influence of tropical cyclones at a specified height are output.

[0049] The technical methods described in this embodiment can be used to establish a high-precision database of tropical cyclone wind fields in marine areas. Based on this database, the following applications can be realized: 1. Extreme Wind Speed ​​Calculation: Simulates each tropical cyclone in the Chinese sea area over the years, outputting the wind speed time series at a specified wind farm location for each tropical cyclone process, serving as the basis for calculating the 50-year return period maximum wind speed. Using the independent storm method and the extreme value type I probability distribution method, the 50-year return period maximum wind speed is calculated for different hub heights (e.g., 100 m, 150 m).

[0050] 2. Macro-level site selection assistance: Draw the impact range of each tropical cyclone that causes severe wind damage in China's sea areas, providing a scientific basis for macro-level site selection and turbine selection for wind farms.

[0051] 3. Input for marine engineering applications: High-precision wind speed sequences are used as driving data for marine hydrological simulations of wind and waves, providing a reliable input field for numerical simulations of extreme waves.

[0052] 4. Floating wind turbine design input: Construct a 30-year long-sequence high-precision wind field dataset of Chinese sea areas to provide technical input for the integrated design of floating wind turbine foundation loads.

[0053] 5. Software System Integration: Develop a software system that integrates functions such as data management, calculation and analysis, visualization, risk assessment, and early warning and forecasting to form a complete solution for tropical cyclone wind field analysis.

[0054] Example 2: This example uses Super Tropical Cyclone "Mangkut" (international designation 202411) as an example to illustrate the implementation process and effects of the tropical cyclone wind field simulation method: S100. Based on the path and impact range of the target tropical cyclone, determine the simulation area and obtain multi-source observation data of the simulation area during the period of the target tropical cyclone.

[0055] Typhoon "Mangkut" formed in the Northwest Pacific Ocean on September 1, 2024 (UTC). Throughout its lifespan, it followed a predominantly westward path, making landfall in Wenchang City, Hainan Province, my country, at 09:00 (UTC) on September 6 and in Quang Ninh Province, Vietnam, at 06:00 (UTC) on September 7. Based on Mangkut's movement path, the simulation scope was determined to be a double-nested system. Domain 1 (parent domain) and Domain 2 (child domain) had resolutions of 9 km and 3 km, respectively. Domain 2 covered Mangkut's movement from 00:00 (UTC) on September 4 to 06:00 (UTC) on September 7. A total of 89 vertical layers were set, with each layer spaced 10 meters apart below a height of 200 meters.

[0056] The physical parameterization scheme is configured as follows: Microphysical processes: the Thompson scheme; Radiation process: RRTMG longwave and shortwave radiation scheme; Boundary layer process: MYNN 2.5th order scheme; Near-surface processes: Revised MM5 Monin-Obukhov scheme; Land surface processes: Noah scheme; Cumulus convection parameterization: Enable the Grell-Freitas cumulus parameterization scheme in Domain 1 and disable it in Domain 2.

[0057] Data source configuration: Driving field: ERA5 reanalysis data (0.25° × 0.25°) Observational data: NCEP ADP observational data Sea surface temperature: OSTIA (Sea Surface Temperature and Sea Ice Analysis) data Observational assimilation data: Data from three lidar systems within Domain 2, with lidar S# used for observational assimilation and lidars D# and T# used for independent verification and evaluation. The S# lidar data was preprocessed into OBS_DOMAIN format, which can be directly read by the real.exe module in WRF mode. The data includes radial and zonal wind components and their corresponding pressure layers at altitudes of 50–200 m.

[0058] S200. The first numerical simulation experiment was constructed using a mesoscale numerical model. The observation approximation assimilation method was used to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results. To optimize the simulation of tropical cyclone intensity, this step designed and executed multiple sets of observation assimilation comparison experiments, the specific design of which is shown in Table 1.

[0059] After conducting the first round of assimilation experiments, based on the CMA-STI optimal path data, the simulation errors of each experiment for the minimum pressure (Pmin) and maximum wind speed (Vmax) at the cyclone center were evaluated. The evaluation results are as follows: Figure 2 As shown, the EXP_ON3 experiment performed best in simulating Pmin and Vmax, so it was determined as the optimal result for the first round of assimilation and denoted as EXP_ON_BEST.

[0060] S300. Based on the first numerical simulation experiment, a second numerical simulation experiment is constructed. Using the spectral relaxation assimilation method, the spectral relaxation parameter configuration is determined based on the first wind field simulation results. The large-scale circulation field in the model is then used to drive data relaxation through the large-scale reanalysis to obtain the second wind field simulation results.

[0061] This step aims to optimize path simulation by testing different spectral relaxation parameters. Based on the obtained EXP_ON_BEST, spectral relaxation assimilation is enabled (only enabled in Domain 1), and multiple sets of parameter sensitivity experiments are constructed to test different spectral relaxation parameter configurations. The specific experimental design is shown in Table 2.

[0062] After conducting the second round of assimilation experiments, based on the optimal path data from CMA-STI, the simulation errors of each experiment regarding cyclone Pmin, Vmax, and path were comprehensively evaluated. The evaluation results are shown below. Figure 2The EXP_ON_SN5 experiment showed the best overall performance in terms of Pmin, Vmax, and path, so it was determined as the final solution and denoted as EXP_ON_SN_BEST.

[0063] S400. Based on the second wind field simulation result EXP_ON_SN_BEST, generate tropical cyclone wind field simulation data.

[0064] To verify the accuracy of EXP_ON_SN_BEST, experimental results were verified using actual LiDAR measurement data. Figure 3 The optimal simulation scheme EXP_ON_SN_BEST was presented, demonstrating the time series of wind speed and wind direction at 100 m and 150 m altitudes, respectively. These simulations were compared with measured data from three lidar stations (S#, D#, and T#), the control group EXP_CTRL, and the reanalysis data ERA5. Data from D# and T# were not included in the assimilation experiment and were independent observations. The results show that EXP_ON_SN_BEST accurately reproduces the temporal variations in wind speed and direction at stations S#, D#, and T#, and its simulation results are closer to the measured values ​​than those from EXP_CTRL and ERA5. However, EXP_CTRL systematically overestimated wind speed during Typhoon Mangkhut's passage, while the ERA5 data completely failed to capture the peak wind speed observed by the lidar during Mangkhut's influence.

[0065] The EXP_ON_SN_BEST experimental results were further verified using wind speed data measured at altitudes of 50–200 m using three lidar sensors (S#, D#, and T#). Statistical verification results for each altitude level (mean absolute error, bias, root mean square error, and correlation coefficient) are as follows: Figure 4 As shown in the figure. The results show that at all altitude levels, the EXP_ON_SN_BEST scheme is significantly better than EXP_CTRL and ERA5.

[0066] Based on the validated EXP_ON_SN_BEST, simulated tropical cyclone wind field data are generated, such as grid point wind speed, wind direction time series data, vertical variation, and spatial distribution map at hub height levels (e.g., 100m, 150m).

[0067] Example 3: This example is a tropical cyclone wind field simulation device that integrates observation and spectral relaxation assimilation, specifically including: The data acquisition module is used to determine the simulation area based on the path and influence range of the target tropical cyclone, and to acquire multi-source observation data during the target tropical cyclone. The multi-source observation data includes at least large-scale reanalysis driving data and wind field observation data within the wind turbine hub height range. The first assimilation simulation module is used to construct the first numerical simulation experiment using a mesoscale numerical model. It uses the observation approximation assimilation method to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results. The second assimilation simulation module is used to construct a second numerical simulation experiment based on the first numerical simulation experiment. It adopts the spectral relaxation assimilation method, determines the spectral relaxation parameter configuration based on the first wind field simulation results, and drives the data relaxation from the large-scale circulation field in the model to the large-scale reanalysis to obtain the second wind field simulation results. The simulation data generation module is used to generate tropical cyclone wind field simulation data based on the second wind field simulation results; The mesoscale numerical model contains at least horizontally nested parent and child domain computational grids in the horizontal direction; the spectral relaxation assimilation is only enabled in the parent domain.

[0068] Example 4: This example is a storage medium that stores a computer program that can be executed by a processor. When the computer program is executed, it implements the steps of the tropical cyclone wind field simulation method described in Example 1.

[0069] Example 5: This example is a tropical cyclone wind field simulation device, which has a memory and a processor. The memory stores a computer program that can be executed by the processor. When the computer program is executed, it implements the steps of the tropical cyclone wind field simulation method described in Example 1.

[0070] Example 6: This example is a method for interpolating wind measurement data in a wind farm, specifically including the following steps: 1. Obtain information on missing measurement periods for the wind measurement towers of the target wind farm, as well as the actual wind measurement data for adjacent non-missing measurement periods.

[0071] 2. Based on the second wind field simulation results obtained by the tropical cyclone wind field simulation method described in Example 2, simulated wind speed time series data corresponding to the location of the wind measuring tower are generated.

[0072] 3. Calculate the correlation coefficient between the measured wind data of the wind measuring tower during non-missing periods and the simulated wind speed time series data.

[0073] Fourth, when the correlation coefficient is greater than a preset threshold, the data for the missing measurement period is interpolated based on the simulated wind speed time series data.

[0074] In this embodiment, the wind speed measurement was missing after Typhoon Capricorn passed by the T# radar (e.g., Figure 3 c, 3f, 3i). The wind measurement sequence at position T# of the optimal simulation scheme EXP_ON_SN_BEST is used as the wind measurement data interpolation product (e.g., Figure 3 c, 3f, 3i (red lines), statistically verified (e.g.) Figure 4The wind measurement data interpolation product generated by this invention more closely matches the wind speed change trend of the T# radar measured data, with a correlation of 0.981 (see...). Figure 4 The correlation between the measured effective data of S# and D# radar and T# radar (0.971 and 0.934, respectively) is higher than that between the control group EXP_CTRL and the reanalysis data ERA5 data and the measured effective data of T# radar (0.954 and 0.830, respectively), which meets the requirements of existing specifications for interpolation of missing wind speeds.

[0075] Example 7: This example is a method for assessing tropical cyclone gale hazards in offshore wind power projects, specifically including the following steps: I. Based on the second wind field simulation results obtained by the tropical cyclone wind field simulation method described in Example 2, spatial distribution data of the maximum wind speed of tropical cyclones at the wind turbine hub height layer are generated.

[0076] II. Compare the spatial distribution data of the maximum wind speed with the preset wind speed threshold for the wind turbine design level; based on the comparison results, identify high-risk disaster areas where the wind speed exceeds the standard.

[0077] In this embodiment, the maximum wind speed at a height of 100 m at each grid point during the impact of Typhoon "Mangkut" was extracted and compared with the IEC design wind speed threshold to draw a risk area map (see...). Figure 5 The maximum wind speed at a height of 100m along Typhoon "Mangkut's" path could reach 73 m / s, exceeding the IECI category (50 m / s). This wind speed range is distributed across the waters east of Hainan Island, the northwestern coastal area of ​​Hainan Island, and the central Beibu Gulf. When developing wind power in these locations, targeted extreme wind speed verification and specific risk assessments should be conducted. Customized designs or the selection of specialized turbine models with higher wind resistance ratings (such as those using Category S design) should be employed, ensuring that their design wind speed is higher than the wind speed value revealed in this simulation.

[0078] If the aforementioned functions 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 this invention, or the part that contributes to the prior art, or a portion 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, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this 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.

[0079] 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-including 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.

[0080] More specific examples (a non-exhaustive list) of computer-readable media 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 aforementioned program can be printed, because the aforementioned 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.

[0081] 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.

[0082] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0083] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0084] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for simulating tropical cyclone wind fields by integrating observation and spectral relaxation assimilation, characterized in that, include: Based on the path and impact range of the target tropical cyclone, the simulation area is determined, and multi-source observation data during the target tropical cyclone are obtained. This multi-source observation data includes at least large-scale reanalysis driving data and wind field observation data within the wind turbine hub height range. The first numerical simulation experiment was constructed using a mesoscale numerical model. The observation approximation assimilation method was used to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results. Based on the first numerical simulation experiment, a second numerical simulation experiment is constructed. The spectral relaxation assimilation method is adopted. Based on the first wind field simulation results, the spectral relaxation parameter configuration is determined. The large-scale circulation field in the model is driven to the large-scale reanalysis to relax the data, and the second wind field simulation results are obtained. Based on the second wind field simulation results, tropical cyclone wind field simulation data are generated; The mesoscale numerical model contains at least horizontally nested parent and child domain computational grids in the horizontal direction; the spectral relaxation assimilation is only enabled in the parent domain.

2. The tropical cyclone wind field simulation method based on the fusion of observation and spectral relaxation assimilation as described in claim 1, characterized in that, The multi-source observation data also includes large-scale integrated meteorological observation data and / or high-resolution sea surface temperature data.

3. The tropical cyclone wind field simulation method based on the fusion of observation and spectral relaxation assimilation as described in claim 1, characterized in that, The method of assimilating multi-source observation data into a mesoscale numerical model using observation approximation includes: Based on the configured mode, a set of observational approximation assimilation experiments is constructed. This set of experiments includes at least a control group and multiple assimilation experiments, which are constructed by adjusting the combination of assimilation data and assimilation parameters. After running this set of experiments, the simulation accuracy of path error, minimum central pressure, and maximum near-center wind speed of each experimental result was comprehensively evaluated based on reliable tropical cyclone optimal path and intensity observation datasets. Based on the evaluation results, the experiment that best simulated the intensity of tropical cyclones was selected from this group of experiments and used as the first wind field simulation result.

4. The tropical cyclone wind field simulation method based on the fusion of observation and spectral relaxation assimilation as described in claim 1, characterized in that, The method employing spectral relaxation assimilation, based on the first wind field simulation results, determines the spectral relaxation parameter configuration, and relaxes the large-scale circulation field of the mesoscale numerical model towards the large-scale reanalysis-driven data, including: Based on the simulation results of the first wind field, the configuration of spectral relaxation assimilation parameters for multiple sets of comparative experiments was determined, including assimilation variables and relaxation coefficients; After running this set of experiments, the improvement effect of each experiment on path error and intensity maintenance capability was evaluated based on reliable tropical cyclone optimal track and intensity observation datasets. Based on the evaluation results, the experiments that can minimize path simulation errors while maintaining or further improving intensity simulation accuracy were selected from this group of experiments and used as the second wind field simulation results.

5. The tropical cyclone wind field simulation method based on the fusion of observation and spectral relaxation assimilation as described in claim 4, characterized in that, The assimilation variables include at least the horizontal wind field component and the temperature field.

6. The tropical cyclone wind field simulation method based on the fusion of observation and spectral relaxation assimilation according to claim 4, characterized in that, The spectral relaxation assimilation parameter configuration also includes the spectral filter cutoff wavelength, which is dynamically determined based on the parent domain range and the Rossby deformation radius of the target tropical cyclone.

7. A tropical cyclone wind field simulation device that integrates observation and spectral relaxation assimilation, characterized in that, include: The data acquisition module is used to determine the simulation area based on the path and influence range of the target tropical cyclone, and to acquire multi-source observation data during the target tropical cyclone. The multi-source observation data includes at least large-scale reanalysis driving data and wind field observation data within the wind turbine hub height range. The first assimilation simulation module is used to construct the first numerical simulation experiment using a mesoscale numerical model. It uses the observation approximation assimilation method to assimilate multi-source observation data into the mesoscale numerical model to obtain the first wind field simulation results. The second assimilation simulation module is used to construct a second numerical simulation experiment based on the first numerical simulation experiment. It adopts the spectral relaxation assimilation method, determines the spectral relaxation parameter configuration based on the first wind field simulation results, and drives the data relaxation from the large-scale circulation field in the model to the large-scale reanalysis to obtain the second wind field simulation results. The simulation data generation module is used to generate tropical cyclone wind field simulation data based on the second wind field simulation results; The mesoscale numerical model contains at least horizontally nested parent and child domain computational grids in the horizontal direction; the spectral relaxation assimilation is only enabled in the parent domain.

8. A storage medium storing a computer program executable by a processor, characterized in that, When the computer program is executed, it implements the steps of the tropical cyclone wind field simulation method according to any one of claims 1 to 6.

9. A tropical cyclone wind field simulation device, comprising a memory and a processor, wherein the memory stores a computer program executable by the processor, characterized in that, When the computer program is executed, it implements the steps of the tropical cyclone wind field simulation method according to any one of claims 1 to 6.

10. A method for interpolating wind measurement data in a wind farm, characterized in that, include: Obtain information on missing measurement periods for the wind measurement towers of the target wind farm, as well as the actual wind measurement data for adjacent non-missing measurement periods; Based on the second wind field simulation result obtained by the tropical cyclone wind field simulation method according to any one of claims 1 to 6, simulated wind speed time series data corresponding to the location of the wind measuring tower are generated. Calculate the correlation coefficient between the measured wind data from the wind measurement tower during non-missing periods and the simulated wind speed time series data; When the correlation coefficient is greater than a preset threshold, the data for the missing time period is interpolated based on the simulated wind speed time series data.

11. A method for assessing tropical cyclone gale hazards in offshore wind power projects, characterized in that, include: Based on the second wind field simulation results obtained by the tropical cyclone wind field simulation method according to any one of claims 1 to 6, spatial distribution data of the maximum wind speed of tropical cyclones at the wind turbine hub height layer are generated. The spatial distribution data of the maximum wind speed is compared with the preset wind speed threshold for the wind turbine design level; Based on the comparison results, high-risk areas for disasters caused by excessive wind speeds were identified.

Images