Simulation and analysis method of floating wind turbines in wind and wave environment considering atmospheric stability

Abstract

The present invention relates to the field of floating wind turbine load simulation, and in particular to a method for simulating and analyzing floating wind turbines in a wind and wave environment taking into account atmospheric stability, comprising: WRF wind field data simulation; WRF‑SWAN wave field simulation; calculation of key parameters for wind field simulation; WRF‑TurbSim wind field simulation; and OpenFAST simulation and analysis. Based on traditional wind field drive, the present invention introduces the Richardson number calculation and atmospheric stability classification methods, combines spectrum inversion with adaptive fitting mechanisms, extracts key parameters of the engineering wind speed spectrum model, and establishes a parameter mapping relationship between WRF output and TurbSim input. By developing a Richardson number-driven disturbance construction module in TurbSim, the adaptive generation of wind speed disturbances under multiple stability conditions is achieved, and the physical consistency and engineering adaptability of the simulated wind speed field in cross-sectional structure and spectral characteristics are improved. Finally, the OpenFAST platform is combined to complete the load analysis of floating wind turbines under wind-wave coupling input, supporting system response evaluation and structural design optimization under complex sea conditions.

Description

Technical Field

[0001] The present invention relates to the field of floating wind turbine load simulation, and in particular to a floating wind turbine simulation analysis method in a wind and wave environment taking into account atmospheric stability, which is mainly used for the dynamic response analysis of floating wind turbine structures under complex sea conditions. Background Art

[0002] Floating Offshore Wind Turbines (FOWTs) are considered the most promising technology for developing deep-sea wind energy resources. However, compared to onshore or offshore fixed wind turbines, floating wind turbines are deployed in harsher and more complex marine environments for long periods of time, and must withstand the combined effects of multiple environmental loads such as wind, waves, and currents. Their supporting platforms exhibit six degrees of freedom: translation (surge, sway, and heave) and rotation (roll, pitch, and bow). This platform motion is strongly coupled with the aeroelastic behavior of the wind turbine itself, the dynamic response of the control system, and the constraints of the mooring system. This complex multi-body dynamics problem makes accurately predicting the structural loads, motion response, fatigue damage, and ultimate limit states of floating wind turbines under actual operating conditions a highly challenging and critical technical problem.

[0003] Therefore, accurate and efficient numerical simulation analysis of the coupled dynamic response of floating wind turbines under the real and dynamic combined action of wind and waves is of vital importance for their conceptual design, performance optimization, structural safety assessment, control strategy formulation and reliable operation and maintenance throughout their life cycle. Traditional industrial analysis methods often use simplified models, such as static or uniform wind fields, regular waves or parameterized wave spectrum models, which may not be able to fully capture the spatiotemporal evolution characteristics of wind and wave fields and their complex interactions with floating platforms and wind turbines, resulting in insufficient prediction accuracy of the dynamic response of the system. In order to improve the fidelity of the simulation and increase the calculation rate as much as possible, it is necessary to develop computational methods that can effectively link high-precision environmental simulation and structural dynamics analysis, and minimize the computational cost while decoupling complex multi-dimensional dynamic problems. This usually involves the use of professional numerical models from different fields:

[0004] (1) Mesoscale meteorological models (such as the Weather Research and Forecasting model, WRF) are used to generate unsteady and non-uniform wind field data with high temporal and spatial resolution.

[0005] (2) The third generation of wave models (such as Simulating Waves Nearshore, SWAN) is used to simulate the generation, propagation and evolution of wind-generated waves and obtain refined wave field parameters.

[0006] (3) Wind turbine integrated simulation tools (such as OpenFAST) are used to couple and solve the aerodynamic, hydrodynamic, control system dynamics, and structural elastic dynamics equations of the wind turbine, simulating the response of the wind turbine system to given environmental inputs.

[0007] Especially under complex climate and ocean conditions, the turbulence spectrum model currently used in load calculations cannot fully capture the temporal and spatial evolution characteristics; the interaction between wind and waves has an important impact on the dynamic response of floating wind turbines, but traditional models may not be able to capture this process well.

[0008] 1) For example, Chinese patent application number 201510400100.8 discloses a wind farm simulation method that couples the WRF and OpenFOAM models. This coupled calculation downscales WRF's kilometer-level horizontal resolution data to OpenFOAM's 30-meter resolution. This method improves wind farm simulation in complex terrain, thereby enhancing the accuracy and performance of wind resource assessment and wind power forecasting.

[0009] 2) For example, patent application number 202410547852.6 discloses an integrated OpenFAST-OpenFOAM coupled calculation method for floating wind turbines. This method implements OpenFAST and OpenFOAM coupled calculations, facilitates the interaction of floating wind turbine body motion data and tower base loads, significantly reduces the runtime of full-scale CFD analysis, and takes into account control and flexible structural deformation under normal power generation conditions.

[0010] But the above technique:

[0011] (1) The focus is mainly on wind farm simulation, without in-depth consideration of wave characteristics and their interaction with wind farms. This limits the ability to assess overall meteorological conditions in complex tidal environments (such as offshore wind farms), and the results may not fully reflect the actual wind energy resources.

[0012] (2) It focuses on the analysis of the floating body movement and tower base load of floating wind turbines. Although there has been some improvement in the dynamic response calculation, it still focuses on the local analysis of wind turbine operation and lacks a systematic analysis of the combined impact of wind and wave fields.

[0013] In summary, existing patented technologies are insufficient to comprehensively assess the wind and wave characteristics of offshore wind farms, resulting in insufficiently in-depth analysis of complex meteorological conditions. Furthermore, the two aforementioned patents focus on local simulations, lacking high-temporal resolution for instantaneous wind speed and wave interactions. Summary of the Invention

[0014] The present invention aims to overcome the shortcomings of the prior art by providing a method for simulating and analyzing floating wind turbines in wind-wave environments that considers atmospheric stability. This method, based on the OpenFAST-WRF-SWAN platform, enables detailed simulation of the load response of floating wind turbines in wind-wave coupled environments. Building on traditional meteorological data-driven approaches, this method incorporates stability identification, spectral inversion, and disturbance reconstruction mechanisms to improve the adaptability and accuracy of wind speed fields under multiple stability conditions.

[0015] To achieve the above object, the present invention adopts the following technical solutions:

[0016] A simulation analysis method for floating wind turbines in a wind and wave environment considering atmospheric stability includes:

[0017] (1) WRF wind field data simulation;

[0018] A multi-layer nested grid structure is established, centered around the wind farm location, for WRF simulations. The innermost grid resolution is 1 km. Bidirectional nesting and multi-physics parameterization are employed, and ground-based wind tower or radar data are combined for simulation bias correction. High-resolution, layered wind speed and temperature profile data are output.

[0019] (2) WRF-SWAN wave field simulation;

[0020] After quality control of the wind field data output by WRF, it is used as the SWAN driver. Multi-spectral and multi-directional resolution settings are introduced. The Westhuysen whitening model and Komen dissipation mechanism are used to model the wave process. The wave height, period and directionality parameters are output, and finally an OpenFAST-compatible wave file is generated.

[0021] (3) Calculation of key parameters for wind field simulation;

[0022] Based on the output results of the WRF model, this section constructs the key physical quantity extraction and processing process required for wind speed field synthesis. By introducing the Richardson number calculation and atmospheric stability classification method, the identification of atmospheric states at different altitudes is achieved. Combined with the fast Fourier transform (FFT) of the time series, the wind speed power spectral density (PSD) is inverted, and the key parameters of spectral types such as Kaimal and von Karman are extracted through the spectral function adaptive fitting technology. Furthermore, the wind field information of the local area where the wind turbine is located can be extracted using a spatial mask, and a high-resolution wind speed profile can be obtained through vertical interpolation, or a wind speed profile can be constructed through an empirical function. The above process establishes a parameter mapping relationship between the WRF simulated wind field and the engineering wind speed spectrum model, providing basic data support for the construction of the disturbance profile and the generation of input files.

[0023] In summary, this link forms a closed-loop system from multi-source WRF variables to wind field structure identification, stability judgment and spectral parameter inversion, establishing a traceable and physically driven data foundation for TurbSim wind speed synthesis.

[0024] (4) WRF-TurbSim wind field simulation;

[0025] Based on the TurbSim source code framework, a modular extension function for WRF output drivers has been developed. By reconstructing the main program structure and input file parsing logic, the automatic reading and dynamic setting of parameters such as wind speed spectrum model selection, wind speed profile calculation, and disturbance superposition are achieved. The newly added disturbance construction module supports user input of Richardson number-driven disturbance terms, which are used to superimpose turbulence characteristics representing different atmospheric stable states on the basis of existing mean profiles. In addition, the adjusted output program can generate .bts format wind speed field files compatible with OpenFAST, ensuring the physical consistency of the generated data in terms of time series, profile structure, and spectral characteristics. This module significantly improves the efficiency of WRF-TurbSim linkage simulation and expands the applicability and engineering versatility of TurbSim under non-standard input conditions.

[0026] (5) OpenFAST simulation and analysis;

[0027] Configure the required wind and wave file paths in OpenFAST, set them to binary TurbSim format, and run the simulation. Perform subsequent data analysis using MATLAB or PyDatView to support wind power project design optimization.

[0028] The beneficial effects of this invention are as follows: first, the wind field is calculated using WRF, then the WRF wind field results are used to drive the SWAN wave field calculation. Finally, the WRF wind field information and the SWAN wave field information are input into OpenFAST to perform a full-system coupled dynamic response analysis of the floating wind turbine. By integrating specialized numerical models from different physical fields and establishing a specific data transmission path, this method aims to provide a more accurate and reliable technical means for refined performance evaluation and design verification of floating wind turbines in complex wind and wave environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of the present invention;

[0030] Figure 2 It is a structural block diagram of the present invention;

[0031] Figure 3 The power spectrum density comparison of different wind spectrum models under different atmospheric stability is used in the present invention;

[0032] Figure 4 The present invention uses a comparison of different wind profiles and disturbance models under different atmospheric stabilities;

[0033] Figure 5 A schematic diagram of the perturbation function used in the present invention;

[0034] Figure 6 The velocity comparison in the u component direction using different wind spectrum models is used in the present invention;

[0035] Figure 7 The present invention uses different wind spectrum models to compare the speed in the v component direction;

[0036] Figure 8 The present invention uses different wind spectrum models to compare the speed in the w component direction;

[0037] Figure 9 The comparison of power generation under the above wind speed using different wind spectrum models in the present invention is shown;

[0038] Figure 10 The present invention uses different wind spectrum models to compare the rotor speed at the above wind speed;

[0039] Figure 11 The present invention uses different wind spectrum models to compare the front and rear displacements of the nacelle at the above wind speeds;

[0040] Figure 12 The present invention uses different wind spectrum models to compare the out-of-plane displacement of the blade at the above wind speed;

[0041] Figure 13 The present invention uses different wind spectrum models to compare the blade surface displacement at the above wind speed;

[0042] Figure 14 The present invention uses different wind spectrum models to compare the longitudinal surge of the floating platform under the above wind speed;

[0043] Figure 15 The present invention uses different wind spectrum models to compare the pitching of the floating platform under the above wind speed;

[0044] Figure 16 This is a comparison of the yaw of the floating platform under the above wind speed using different wind spectrum models according to the present invention. DETAILED DESCRIPTION

[0045] The present invention will be further described below with reference to the accompanying drawings and examples.

[0046] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for understanding and reading by those familiar with this technology. They are not used to limit the conditions for implementation of the present invention and therefore have no substantial technical significance. Any modification of the structure, change in the proportional relationship, or adjustment of the size should still fall within the scope of the technical content disclosed by the present invention without affecting the efficacy and purpose of the present invention. At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" quoted in this specification are only for the convenience of description and are not used to limit the scope of implementation of the present invention. Changes or adjustments in their relative relationships should also be regarded as the scope of implementation of the present invention without substantially changing the technical content.

[0047] like Figures 1-16 As shown in the figure, the simulation analysis method of floating wind turbines in wind and wave environment considering atmospheric stability includes:

[0048] 1. WRF wind field data simulation;

[0049] (1). Nested domain settings;

[0050] Step 1. Define the target area: Determine the site for the wind farm and obtain its precise latitude and longitude. For example, if the site is in a specific city or region, you can use a mapping service (such as Google Maps or Amap) to retrieve its latitude and longitude information.

[0051] Step 2. Set the domain boundary: Determine the boundary of the innermost domain. Usually, the range that the wind farm may be affected should be considered when selecting the domain boundary. Using the extension setting, assuming the radius of the wind farm is 20km, the boundary of the innermost domain can be set to the center longitude and latitude plus or minus 0.18°

[0052] Step 3. Define additional nested domains: Define middle and outer domains. Each domain should overlap with the innermost domain to ensure effective nesting. The inner domain (1km resolution), middle domain (3km resolution), and outer domain (9km resolution) are centered at the wind farm site. The innermost domain is the wind farm site, with a resolution of 1km. Establish a 3:1 resolution ratio between subdomains.

[0053] Step 4. Set the resolution and nesting properties of each domain by configuring the namelist.input file of the WRF model.

[0054] Each domain is identified by its parent_id and grid_id. For example: Outer domain: grid_id = 1, parent_id = 0. Middle domain: grid_id = 2, parent_id = 1. Inner domain: grid_id = 3, parent_id = 2. The number of horizontal and vertical grid points for each domain is represented by e_we and e_sn. Each domain is identified by its parent_id and grid_id. For example: Outer domain: grid_id = 1, parent_id = 0. Middle domain: grid_id = 2, parent_id = 1. Inner domain: grid_id = 3, parent_id = 2. The number of horizontal and vertical grid points for each domain is represented by e_we and e_sn.

[0055] Step 5. Bidirectional Nesting Setup: After ensuring nesting, check the bidirectional nesting parameter settings. In WRF, feedback is used to improve simulation accuracy by allowing the target domain to influence the surrounding domains. Enabling bidirectional nesting ensures that simulation results in the inner domain can be fed back to the outer domain and influence its simulation results.

[0056] Step 6. Grid division and verification: Use NCL or Python libraries (such as matplotlib) to draw the grid settings and key areas of each domain to ensure the coverage and nesting of the domains are accurate.

[0057] (2) WRF preprocessing;

[0058] Step 7. Collect high-resolution terrain data: Download high-resolution data suitable for WRF from the WRF official website, such as USGS, GTOPO30, and other formats. If you need high-precision terrain data for China or a specific region, you can access the Digital Elevation Model (DEM) Data Center of Tsinghua University (such as THUDEM).

[0059] Step 8. Determine the data format: Terrain data is typically in NetCDF format. Ensure the dataset complies with WRF requirements. Check for a valid coordinate system (e.g., WGS84) and ensure the data covers the entire simulation area.

[0060] Step 9. Collect high-resolution meteorological data: Download high-resolution meteorological data (such as ERA5, NARR, GFS, GDAS, etc.) from the National Center for Atmospheric Research (NCAR) website. Download data according to the simulation time period, ensuring that the data frequency is consistent with the simulation requirements (for example, every hour or every three hours).

[0061] Step 10. Weather data processing: Use the ungrib tool to process the downloaded weather data and ensure that the data is converted into a format that WRF can recognize. You can use the following command line operation to decompress the weather data: . / ungrib.exe <input_data。

[0062] Step 11. Use WPS tools: The WRF preprocessing stage includes multiple tools, including geogrid.exe, metgrid.exe, and ungrib.exe. Use geogrid.exe to process terrain data, generating high-resolution terrain and ocean data with the output files geo_em.d0*.nc. Use ungrib.exe to convert meteorological data to WRF format, using the configuration file ungrib.cfg. Use metgrid.exe to combine the initial meteorological data with the processed terrain data, generating the files wrfinput_d0* and wrfbdy_d0*.

[0063] Step 12. Data inspection after preprocessing: Check the generated wrfinput (initial field input) and wrfbdy (boundary field input) files. Use tools such as ncdump or ncview to view the contents of the generated files to ensure that the terrain and meteorological data are accurate. If any data issues are found (such as missing NaN values ​​or unreasonable terrain heights), return to the original data source to ensure that the data download and processing process were correct.

[0064] (3) Sensitivity test;

[0065] Step 13. Determine the purpose of sensitivity testing: Before conducting the final WRF simulation, sensitivity testing is required to determine the optimal parameterization scheme. This step ensures that the selected parameters are appropriate for the simulated wind farm area, thereby improving simulation accuracy. The purpose of this method test is to determine the parameterization combination with the greatest impact on wind farm simulation, including planetary boundary layer parameterization, microphysical process parameterization, and deep convective cumulus parameterization. By validating the simulation results of different schemes, the optimal combination is identified to reduce the deviation between the simulated results of wind speed, wind direction, and other parameters and the observed data.

[0066] Step 14. Select the test parameterization scheme: In the sensitivity test, it is necessary to select a variety of alternative parameterization schemes. The following are the key schemes that need to be considered: Planetary boundary layer (PBL) scheme. Commonly used PBL schemes include YM, MYJ, YSU, etc. The optimal scheme is selected by evaluating their impact on the wind field at different altitude levels; Microphysical scheme. Microphysical process schemes include WSM6, Thompson, Ferrier, etc. By changing the precipitation simulation effect, their impact on atmospheric stability and wind speed is evaluated; Cumulus schemes include Kain-Fritsch, Betts-Miller-Janjic, etc. By analyzing the performance of these schemes under deep convection conditions, their impact on wind field simulation is understood.

[0067] Step 15. Experimental Design: Select experimental combinations based on different parameterization schemes. For each test combination, set the same initial and boundary conditions to ensure comparable results. For example, five different scheme combinations could be designed: PBL: YSU, Microphysics: WSM6, Cumulus: Kain-Fritsch; PBL: MYJ, Microphysics: Thompson, Cumulus: Betts-Miller-Janjic; PBL: YSU, Microphysics: Ferrier, Cumulus: Kain-Fritsch; PBL: MYJ, Microphysics: WSM6, Cumulus: Betts-Miller-Janjic; PBL: YSU, Microphysics: Thompson, Cumulus: Ferrier. Initially, choose a timeframe of 12-24 hours to capture the dynamic nature of wind field evolution.

[0068] Step 16. Set up NAMELIST configuration: Configure different parameterization schemes in the WRF input file namelist.input so that they can be called when running WRF.

[0069] Step 17. Start WRF simulation: For each designed experimental combination, call the WRF main program to simulate.

[0070] Step 18. Output results: Record the simulation results for each test combination, including wind speed, wind direction, and other key meteorological parameters. Generate data at a specified output frequency (e.g., hourly output).

[0071] Step 19. Data Comparison and Results Analysis: Compare the WRF-simulated wind field results with data from wind towers or other meteorological observation stations. Use statistical process control (SPC) metrics such as root mean square error (RMS) and correlation coefficients to evaluate the results. Calculate the deviation of each set of experimental results, find the minimum RMS error and other correlation metrics, and identify the optimal parameterization combination. Record the performance metrics for each combination, compare, and determine the best performing parameter combination. Ensure that the selected combination is recorded in the namelist.input file.

[0072] Table 1, the present invention uses the sensitivity test scheme example (pbl2_mp6_cu11_fdda2);

[0073]

[0074] (4) WRF simulation domain data output settings.

[0075] Step 20: Integrate the parameterization scheme: Before running the WRF atmospheric model simulation, you first need to integrate the optimal parameterization scheme determined in the sensitivity test. This will ensure that the simulation follows the verified settings and improve the accuracy of the results.

[0076] Step 21: Start the WRF simulation: Use the prepared initialization and boundary condition data to obtain atmospheric and sea surface wind data for the target area. Load the wrfinput (initial field input) and wrfbdy (boundary field input) files. Ensure that the file names for the initial and boundary conditions are correctly set in namelist.input so that the model can read them. Set the simulation duration and output frequency: Set the total simulation duration and the frequency of output results (e.g., hourly). In the terminal, navigate to the WRF model runtime directory and invoke the WRF main program, wrf.exe, to run the model.

[0077] Step 22 Output data extraction: After the WRF simulation is completed, it is necessary to extract the sea surface wind field data and other atmospheric parameters, and conduct subsequent data analysis. Confirm the file format of the output results. The default output of WRF is NetCDF format (such as wrfout_d01_YYYY-MM-DD_HH:MM:SS) to ensure direct compatibility with subsequent models (such as SWAN). Extract data at a specific height: Extract the wind speed at 10 meters (U10, V10 components) on the sea surface and wind speed information at other height layers. You can use the xarray library in Python to read and extract data. In order to obtain more accurate wind field simulation results, the WRF simulation output data needs to be interpolated. For example, spatial interpolation, based on the simulation results, interpolate the wind speed data to a higher spatial resolution (such as 0.1 km) using the interpolation function in the scipy library. Secondly, there is time interpolation, which resamples the data to a finer time scale, such as every 10 minutes, through the resampling function of xarray. The calculation method for some of the key variables can be adopted as follows:

[0078] For example, the average wind speed in the target area at height z for

[0079] ,

[0080] in, N is the total sampling time, u(z,t i ) and v(z,t i ) are at height z, and the time step is t i The u and v components of the velocity at .

[0081] Turbulence intensity σ in WRF u wrf The calculation of (z) first requires the calculation of wind speed disturbance u'(z,t i ) :

[0082] ,

[0083] in, u(z,t i ) It is t i The wind speed at height z at time, is the average wind speed at height z, u'(z,t i ) It is wind speed disturbance.

[0084] 2. WRF-SWAN wave field simulation;

[0085] Step 1: Check the quality of wind farm data, remove outliers, and use spatial interpolation (such as bilinear interpolation) to match the SWAN grid when necessary.

[0086] Step 2: Define the SWAN computing domain;

[0087] First, set the grid range (using WRF latitude and longitude or UTM coordinates) and resolution (recommended ≤ 1 km, nearshore resolution ≤ 500 m). Configure the time parameters: total simulation duration (recommended 3600 seconds or longer) and time step (the wind field drive step should be synchronized with the WRF output, such as 10 minutes. The wave calculation step is recommended to be 60-300 seconds to meet the CFL condition). For grid type selection, refer to the following method: Curvilinear grids are preferred for nearshore areas, and the CGRIDCURV command is used to define the curved coastline fit. Rectangular grids can be used in deep-sea areas, and CGRID REG is used to simplify the calculation.

[0088] Step 3: Physical process configuration;

[0089] Activate the Westhuysen nonlinear wave interaction model (GEN3 WESTHUYSEN) to simulate wave energy transfer. For details, refer to the Komen White Cap Dissipation Model (WCAP KOMEN), set the shallow water wave breaking rate, and use the Collins bottom friction formula (FRICTION COLLINS).

[0090] Step 4: Data correction;

[0091] A correction factor of 1.05 can be applied to the WRF wind speed to compensate for underestimation of ocean wind speeds. A 5° wind direction offset can be added to correct for systematic deviations in the WRF wind direction.

[0092] Step 5: Boundary condition setting;

[0093] For incident wave boundary conditions, consider JONSWAP spectral simulation. Hydrological parameters at the wind farm site are extracted to define spectral parameters (such as wave height and peak period) to reflect local ocean conditions. For strong winds and waves, use the Pierson-Moskowitz spectrum. Alternatively, open boundary conditions can be used to minimize wave reflections and impacts, or reflecting boundary conditions can be used to simulate nearshore wave behavior.

[0094] Step 5: Output parameter settings;

[0095] Select significant wave height (HSIGN), peak period (TPS), mean wave direction (DIR), wavelength (WLEN), and phase propagation as output parameters. Format the output as NetCDF for easy reading by OpenFAST.

[0096] 3. Calculation of key parameters for wind field simulation:

[0097] Step 1: Open the WRF output file.

[0098] Use the netCDF4 library to open the output file generated by the WRF model to read the required meteorological variables. This step ensures that the file is opened in read-only mode so that the data can be read safely later.

[0099] Step 2: Set the time range.

[0100] Define the data period to be extracted by setting the time index slice. For example, you can select multiple time periods, such as 36 to 44, to meet specific project requirements.

[0101] Step 3: Define meteorological variables.

[0102] Key meteorological variables in WRF output, including potential temperature (theta), vertical wind speed, horizontal wind speed (u and v components), ground height (z), surface wind speed, and boundary layer height (PBLH), were extracted for subsequent analysis.

[0103] Step 4: Set the geographic scope.

[0104] Set the geographic boundaries of the analysis area and define the area of ​​interest by specifying the minimum and maximum latitude and longitude. This can be achieved by using spatial masking operations using geographic grid data.

[0105] Step 5: Interpolation calculation.

[0106] At a given height level (e.g., 90 meters for the turbine hub and 300 meters for the top of the wind farm grid), the values ​​of the extracted meteorological variables at that height are calculated using an interpolation method. This step ensures that the required meteorological parameters are obtained at a specific height for wind energy simulation in OpenFAST. Interpolation is performed using linear interpolation:

[0107]

[0108] in, p(z) is the value of the air pressure at height z to be calculated, p bottom is the bottom z of a specific altitude layer bottom The corresponding variable value, p top is the top of a specific altitude layer ztop The corresponding variable value, z, is the specific height that needs to be interpolated, between the two known heights.

[0109] Step 6: Apply spatial masking.

[0110] Perform spatial masking on the extracted results to focus on the meteorological parameters in the area of ​​interest and exclude values ​​outside the specified area. This can be achieved by Boolean masking. Suppose there is a meteorological parameter P (here we use air pressure as an example, but it can also be variables such as temperature and humidity) in a two-dimensional array with the dimensions of (i, j ), corresponding to the rows and columns of the model grid. Define a Boolean mask M, which is also an array of the same dimension, indicating whether each point is in the region of interest. Mask M(i, j) and the extracted pressure value P after applying the mask masked (i, j) They are:

[0111] .

[0112] Step 7: Calculate important meteorological parameters.

[0113] The extracted meteorological parameters are used to calculate the important indicators of gradient Richardson number and fluctuating stress. Gradient Richardson number is used to assess atmospheric stability, while fluctuating stress is used to analyze the impact of wind speed fluctuations on meteorological processes. Specific calculations include:

[0114] The gradient Richardson number formula is applied at a specific altitude to evaluate the airflow characteristics and is as follows:

[0115] ,

[0116] Where g is the acceleration due to gravity; θ is the neutral temperature at the measurement height, in K, which can be derived from T, P, and PB in WRF, where T is the perturbed potential temperature, P is the perturbed air pressure, and PB is the ground state air pressure; u,v are the east-west and north-south wind speeds, in m / s; z is the height, obtained from PH and PHB of WRF, where PH is the disturbance potential and PHB is the ground potential; and , is the rate of change of the neutral temperature and velocity component with height, which is calculated by the central difference approximation;

[0117] Fluctuation stress (PC_UW, etc.) is calculated by taking the pulsation value of each wind speed component and calculating its mean square value. The formula is as follows:

[0118]

[0119] in, PC_UW It is hoped that the u' 、 w' Mean Reynolds stress, in m 2 / s 2 , u' j,correlated yes j Point u The associated value of the directional wind speed, which is independent of the u direction wind speed u' j,independen Data is equal; v' j,correlated yes j Point v The associated value of the directional wind speed is obtained by u、v、w Independent wind speed in direction u' j,independen 、 v' j,independen and w' j,independen The data is combined and calculated; similarly, w' j,correlated yes j Point w The associated value of the direction wind speed is obtained by u' j,independen and w' j,independen The data is combined and calculated. a uv 、a vw 、a uw is the correlation coefficient, ensuring the correlation between different wind speed variables, U' hub,correlated and W' hub,correlated Refers to the corresponding correlation values ​​of the u component and w component of the velocity at the fan hub.

[0120] Step 8: Classification criteria for atmospheric stability states;

[0121] Based on a large number of LES simulations and field observations, the present invention adopts the following classification scheme:

[0122]

[0123] Step 9: Wind spectrum inversion and parameter fitting;

[0124] Based on the wind speed spectrum inversion and construction method output by WRF wind field, the spectral characteristics of local wind speed disturbances are extracted by performing fast Fourier transform (FFT) on the WRF simulation results. This spectral information is then embedded into the TurbSim input parameters, thereby achieving the generation of user-defined wind speed spectra that are different from the standard IEC spectrum, and improving the simulation accuracy of floating wind turbines under non-IEC wind conditions.

[0125] Step 9.1: Extract WRF wind speed time series data;

[0126] Select the central grid of the target wind farm area (or several representative grid points in the area) and extract the horizontal wind speed at the rotor height (e.g. 120m) u(t) Time series, sampling time interval is Δ t , a total of N Sample points form a discrete wind speed sequence:

[0127] ,

[0128] U 0 ,u 1 ,…,U N-1 Represents the wind speed at different time steps;

[0129] The sampling frequency is: , the corresponding maximum frequency (Nyquist frequency) is .

[0130] Step 9.2. Calculate wind speed power spectrum density (PSD) using FFT.

[0131] For a certain height wind speed sequence u(n),(n=0,1,…,N-1) Perform a discrete Fourier transform:

[0132] ,

[0133] in, k is the frequency index, the actual frequency , f s is the sampling frequency, N is the number of sample points.

[0134] Then calculate its one-sided power spectral density (PSD):

[0135] ,

[0136] Step 9.3: Spectral model selection and fitting;

[0137] According to the "local stability state", the spectrum model is initially selected, for example:

[0138] In highly turbulent states, such as S1~S2, the von Karmon spectrum is used.

[0139] ,

[0140] in, S u (f) is the longitudinal wind speed power spectral density, σ u spec is the standard deviation of the wind spectrum, f is the frequency, L u is the integral scale length, which will be fitted by WRF results later. U is the average wind speed.

[0141] In relatively neutral and weak atmospheric states, such as S3~S4, Kaimal spectrum is used.

[0142] ,

[0143] in, S u (f) is the longitudinal wind speed power spectral density, σ u spec is the standard deviation of the wind spectrum, f is the frequency, L u is the integral scale length, which will be fitted by WRF results later. U is the average wind speed.

[0144] In a strongly stable atmospheric state, such as S5, a smooth spectrum is used, which is as follows:

[0145] ,

[0146] This method can adaptively and smoothly express the changing trend of the actual PSD without forcibly aligning the standard spectrum model. a is the amplitude coefficient of the power law spectrum, and its dimension varies with b And change, b is the slope exponent of the power law spectrum (dimensionless). The above parameters can be obtained by least squares fitting or Bayesian optimization.

[0147] For the wind spectrum of a steadily stratified area, the steady wind can be used directly instead, assuming that it has no turbulence.

[0148] Step 9.4. Fit the wind spectrum using the least squares method;

[0149] Based on the above steps, an error function ε(θ) is created to describe the power spectral density PSD of WRF and the error of the preliminary model.

[0150] ,

[0151] in, r i (θ) is the residual at the frequency point, and the error function is the sum of squares of all residuals. S u WRF ( f i ) is the PSD estimated from the WRF data, S u ( f i ; θ) is the parameterized spectral model value. = (σ u , L u ) is the set of parameters that need to be fitted.

[0152] Here, the initial value of the parameter set is set, θ 0 = (σ (0) u , L (0) u ), σ (0) u is the initial value of the standard deviation in the wind spectrum, L (0) u is the initial value of the integral scale length; where σ (0) u The standard deviation σ is obtained from the numerical value of WRF WRF u consistent.

[0153] in, ， ，

[0154] The value of the integral scale can be obtained by the above formula, is the integral scale parameter, which can be described by the following formula:

[0155] ,

[0156] For the error function ε( θ )To solve the problem, optimization algorithms such as Levenberg-Marquardt, trust region method, and BFGS can be used to find the parameters that minimize the error. In this paper, the Levenberg-Marquardt (LM) algorithm is used as an example because it is more suitable due to its nonlinear least squares, low-dimensional parameter space, and faster convergence.

[0157] LM is a balanced method between Gauss-Newton method and gradient descent method. Its core update formula is:

[0158] ,

[0159] in, J is the Jacobian matrix of the error term to the parameter, T means transposing the matrix, I is the identity matrix, r is the residual vector, and λ is the damping coefficient, which is used to adaptively balance the convergence direction and step size during the iteration process.

[0160] By setting the maximum number of iterations, error tolerance, and repeating the iterations until the error converges or the maximum number of steps is reached, the optimal fitting parameters can be obtained: , σ * u is the standard deviation of the wind spectrum, L * u is the integration scale length.

[0161] Step 10: Profile perturbation coupled spectrum driving module;

[0162] In this section, in order to facilitate engineering applications, the present invention provides two feasible profile perturbation reconstruction methods according to the different WRF resolutions.

[0163] 10.1. Match the wind profile function according to the state;

[0164] When the resolution of WRF is insufficient, the method in this section can be used to reconstruct the profile.

[0165] According to different atmospheric conditions, an adapted profile function is selected for the average wind speed distribution:

[0166] For S1~S2, the logarithmic wind profile can be used:

[0167] ,

[0168] in, is the average wind speed at height z; u *is the friction velocity, which represents the effect of surface friction on wind speed; k is the Karman constant (usually about 0.4), which is an empirical constant in the logarithmic wind profile theory; z0 is the surface roughness length, which represents the parameter of surface irregularity and roughness in the model.

[0169] For the stable boundary layer, the stable modified Monin-Obukhov wind profile is used:

[0170] ,

[0171] Where ψ is the stability correction function and L is the Obukhov length, with a positive value indicating a stable atmosphere and a negative value indicating an unstable atmosphere, estimated from WRF output. is the height z Average wind speed; u * Friction velocity, which indicates the effect of surface friction on wind speed; κ is the Karman constant (usually about 0.4), an empirical constant in logarithmic wind profile theory; z0 is the surface roughness length, a parameter representing the surface irregularity and roughness in the model.

[0172] For strong disturbances, use

[0173] ,

[0174] in, is the height z The average wind speed at U hub is the wind speed at the turbine hub, Z hub is the height of the wind turbine hub, α is the wind speed profile power exponent, and the mathematical expression of the disturbance term ε(z) is:

[0175] ,

[0176] in, f i is the frequency point uniformly sampled in the one-sided PSD, A i (z) is given by the spectral density S( f i ) and the integral height, Φi is a random phase, ranging from [0,2π], t For the total time, specifically, A i The expression of (z) is:

[0177] ,

[0178] S (f i ) is the longitudinal wind speed power spectral density, ∆f is a discrete frequency interval, G (z) is expressed as a vertical scaling function (e.g., exponential decay or amplitude retention), reflecting the weakening or strengthening of turbulence amplitude with height;

[0179] ,

[0180] Z is the desired height, z ref is the reference height, such as hub height; z ABL is the boundary layer thickness, which can be provided by WRF; γ and β are modulation coefficients, which depend on the stability state settings (for example, S1 takes γ=0.3 and β=0.1; S5 can take γ=-0.5 and β is 1.0). This function ensures that: for an unstable layer, the intensity of the disturbance increases with altitude; for a stable layer, the disturbance decays rapidly with altitude.

[0181] 10.2 Directly use the WRF vertical wind speed distribution (such as 90m) through interpolation fitting, and then superimpose the disturbance term. This is a data-driven completion method and is used in scenarios where the WRF data granularity is sufficient.

[0182] 4. TurbSim code modification and operation;

[0183] TurbSim currently only has a few fixed wind speed spectrum models, such as the Kaimal spectrum and von Karman spectrum. If you want to apply the wind spectrum model based on atmospheric stability proposed in this method (for example, different wind spectra corresponding to states S1-S5), you need to add a new wind spectrum model to the TurbSim source code.

[0184] Step 1: Modify the source code to support custom wind spectra and wind profiles

[0185] Locate the existing wind spectrum model and find the section in the TurbSim code that defines the wind speed spectrum model. Add the wind spectrum model (such as von Karman, Kaimal, etc.) and wind profile function based on atmospheric stability described in this method to TurbSim. Develop the corresponding calculation formula for each state type.

[0186] Step 2: Modify the wind speed spectrum selection logic: In the main program of TurbSim, make sure that the given Ri_g (gradient Richardson number) and the wind speed spectrum corresponding to the altitude layer selection.

[0187] Step 3: Add matching function: Check the matching between the wind spectrum model and the Richardson number.

[0188] To ensure that the selected wind spectrum model matches the given Ri_g, you need to add a check function to the code to check whether the wind spectrum model meets the requirements. If it does not match, the program should report an error. During the runtime, this function is called before the wind spectrum model is officially applied.

[0189] Step 4: Wind speed related input file settings;

[0190] Before passing calculated wind speed data and atmospheric state information to TurbSim, you must first ensure that all relevant input files are prepared. Key steps include: configuring the TurbSim input file. In the TurbSim inp input file, set WrADFF to 'TRUE', which outputs full-field time series data in TurbSim / AeroDyn format; setting TurbModel to USRVKM; this configuration instructs TurbSim to use a custom von Karman spectrum for wind speed simulation, rather than the default IEC wind speed spectrum. The von Karman spectrum is widely used to describe turbulence characteristics, reflecting wind speed variations under dynamic and intense meteorological conditions; and specifying wind speed profile data. In the configuration file, specify the ProfileFile parameter. This file should contain the wind speed profile data output by the WRF model. Specifically, the ProfileFile should include meteorological parameters such as wind speed, wind direction, and the standard deviation of wind speed (i.e., turbulence intensity) at each altitude. Each line in the file should be organized in ascending order of altitude to ensure that TurbSim can correctly read and apply the wind speed field data at each altitude level. For example, a line of data in a file might contain the following: Altitude: Indicates the altitude of the data point (unit: meters); Wind Speed: The average wind speed at that altitude level (unit: m / s); Wind Direction: The wind direction at that altitude level (unit: degrees); Wind Speed ​​Standard Deviation: The standard deviation of turbulence intensity (unit: m / s); Length Scale: A parameter describing the size of turbulent eddies, reflecting the spatial extent of turbulence in the wind speed field. This data format helps TurbSim generate a corresponding wind speed spectrum for each specified point at different altitude levels.

[0191] Step 5: Set non-IEC meteorological boundary conditions;

[0192] In the TurbSim configuration file, further set non-IEC meteorological boundary conditions related to the boundary layer state. Specific operations include

[0193] Set the Gradient Richardson Number (Ri_g): The Gradient Richardson Number (Ri_g) is used to characterize the stability of the meteorological layer. In the TurbSim configuration file, you need to provide a Gradient Richardson number for each time step based on the meteorological data output by WRF. This parameter helps describe the stability state of the boundary layer, thereby affecting the fluctuation characteristics of the wind speed field. For example, in strong unstable conditions, the Ri_g value is small, the turbulence is stronger, and the wind speed changes more violently. In a stable layer, the Ri_g value is large, the turbulence is suppressed, and the wind speed changes smoothly.

[0194] Setting the fluctuating stress at hub height: Fluctuating stress is a key parameter describing the intensity of turbulent eddies in the wind velocity field. TurbSim uses the fluctuating stress at hub height to adjust the turbulent characteristics of the wind velocity field. You can specify the fluctuating stress at a specific height (such as hub height) by setting the corresponding parameters in the inp file. This value is generally closely related to the turbulence intensity and needs to be set based on the TKE output from WRF or field measurements.

[0195] Spectral Correction for Custom Boundary Conditions: Depending on the selected wind spectrum model (e.g., Kaimal or von Karman), the frequency bands of the wind field need to be adjusted to ensure that the generated wind speed field is realistic under different atmospheric conditions. This spectral correction allows you to select different wind spectrum types based on the atmospheric state (e.g., strong stability, weak stability), and adjust the turbulence intensity and integral scale parameters.

[0196] Step 6: Run TurbSim to generate wind files;

[0197] Start TurbSim using the configured inp input file. TurbSim will generate the target wind speed field based on the input wind spectrum data, meteorological boundary conditions and disturbance model. During operation, TurbSim will dynamically calculate the wind speed field at each time step based on the preset wind speed model, spectrum type and disturbance characteristics, and finally output a .bts file that meets the requirements of OpenFAST. The output .bts file contains the three-dimensional wind speed field data at each moment, which is suitable for OpenFAST to perform subsequent wind turbine load simulation. Each time point in the file contains information such as wind speed component, turbulence intensity, and wind speed standard deviation. OpenFAST can read and use this data to calculate the structural response of the wind turbine and simulate the load and dynamic response of the wind turbine under different atmospheric conditions.

[0198] Step 7: Check the results;

[0199] After generating the wind file, you can check the generated .bts file using MATLAB or pyDatView to ensure that the wind speed field data meets expectations and can be correctly read into OpenFAST. If necessary, you can adjust the input parameters or wind speed spectrum model for further optimization.

[0200] 5.OpenFAST simulation:

[0201] Step 1: Set the wind file to be used;

[0202] In OpenFAST's InflowWind INPUT FILE, set wind type to 3 = binary TurbSimFF. Also, add the path to the bts file in Parameters for Binary TurbSim Full-Field files.

[0203] Step 2: Set the wave file to be used;

[0204] Set WaveMod to 6 in OpenFAST's hydrodyn.dat and set the wave NetCDF file path in WvKinFile.

[0205] Step 3: Run OpenFAST;

[0206] Open a terminal in the file path and run OpenFAST. You can then use MATLAB or PyDatView to analyze the data.

[0207] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A simulation and analysis method for floating wind turbines in a wind and wave environment taking into account atmospheric stability, characterized by: include: (1) WRF wind field data simulation; Nested domain setting: a three-layer nested region is established, with the wind farm site as the center and the innermost layer as the wind farm location domain. The resolution of the innermost layer domain is 1km, and a bidirectional nesting method is adopted. High-resolution terrain and meteorological data are preprocessed, and the parameterization scheme is optimized in combination with sensitivity testing to ensure that the simulation effect is consistent with the wind tower data. The simulation results are interpolated to obtain high temporal resolution sea surface wind speed and data of different altitude layers; (2) WRF-SWAN wave field simulation; After wind farm data undergoes quality verification and outlier removal, the SWAN model is used to simulate wave fields. The computational domain and grid parameters are defined, and modern physical process models are used to simulate wave energy transfer. By setting appropriate wave spectra and boundary conditions, suitable wave output is generated for subsequent OpenFAST reading. (3) WRF-TurbSim wind field simulation; Based on the WRF simulation results, wind files were generated using TurbSim; Extract meteorological variables, perform interpolation and spatial masking, calculate the gradient Richardson number and key parameters of fluctuating stress, and evaluate wind field stability; Integrate meteorological parameters into TurbSim input files and run them to generate binary wind files for OpenFAST; The specific steps for the key wind field parameters of WRF-TurbSim wind field simulation are as follows: Step 1: Open the WRF output file. Use the netCDF4 library to open the output file generated by the WRF model to read the required meteorological variables; this step ensures that the file is opened in read-only mode so that the data can be read safely later; Step 2: Set the time range, Define the data period to be extracted by setting the time index slice; Step 3: Define meteorological variables, Extract key meteorological variables from WRF output, including potential temperature, vertical wind speed, horizontal wind speed, ground height, surface wind speed, and boundary layer height, for subsequent analysis; Step 4: Set the geographical scope, Set the geographic boundaries of the analysis area and define the area of ​​interest by specifying the minimum and maximum latitude and longitude; Step 5: Interpolation calculation, At the set altitude, the values ​​of the extracted meteorological variables at that altitude are calculated by interpolation. This step ensures that the required meteorological parameters are obtained at a specific altitude for wind energy simulation in OpenFAST. Step 6: Apply spatial mask, The extracted results are spatially masked to focus on the meteorological parameters in the area of ​​interest and to exclude values ​​that are not in the specified area; Step 7: Calculate important meteorological parameters, The extracted meteorological parameters are used to calculate the important indicators of gradient Richardson number and fluctuating stress; gradient Richardson number is used to evaluate atmospheric stability, while fluctuating stress is used to analyze the impact of wind speed fluctuations on meteorological processes; Step 8: Determine the stability state classification criteria based on a large number of LES simulations and field observations; Step 9: Wind spectrum inversion and parameter fitting, By performing a fast Fourier transform on the WRF simulation results, the spectral characteristics of the local wind speed disturbance are extracted and the spectral information is embedded in the TurbSim input parameters. This enables the generation of user-defined wind speed spectra that differ from the standard IEC spectrum, thus improving the simulation accuracy of floating wind turbines in non-IEC wind conditions. Step 10: Two methods for reconstructing profile disturbances are provided, depending on the WRF resolution. One is to match the wind profile function according to the state when the WRF resolution is insufficient. The second is to directly use the WRF vertical wind speed distribution through interpolation fitting and then superimpose the disturbance term, which is used in scenarios with sufficient WRF data granularity. Step 11: Modify and run the TurbSim code. Modify the source code to support custom wind spectra and wind profiles, modify the wind speed spectrum selection logic, and add matching functions: check the matching of the wind spectrum model and the Richardson number. In the TurbSim input file inp, set WrADFF to 'TRUE', TurbModel to USRVKM, and the wind field type to USR; it is necessary to combine the relevant atmospheric parameter files calculated by WRF; specify the Profiler file, which should contain wind speed and related meteorological parameters obtained from WRF output; add wind speed, wind direction, and wind speed standard deviation parameters for each altitude layer as the altitude increases; Step 12: Set non-IEC meteorological boundary conditions, In the TurbSim configuration file, set the gradient Richardson number of the target area and the pulsating stress parameters at the hub height; Step 13: Run TurbSim to generate wind files. Run TurbSim and generate a bts file that can be read by OpenFAST; (4) OpenFAST simulation and analysis; Configure the required wind and wave file paths in OpenFAST, set them to the binary TurbSim format, and run the simulation. Perform subsequent data analysis using MATLAB or PyDatView to support wind power project design optimization.

2. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: In the nested domain setting in (1), the resolution ratio between each subdomain is 3:1, that is, the resolution between each layer is 9km:3km:1km.

3. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: In (1), the resolution and nesting properties of each domain are set by configuring the namelist.input file of the WRF model, and parent_id and grid_id are used to identify each layer of domain.

4. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: In (2), the computational domain is defined. First, the grid range and resolution are set, and the time parameters are configured: the total simulation time and the time step. The grid type selection method is: Curvilinear grid is used in the nearshore area, and the curved coastline fitting is defined by the CGRID CURV command; Rectangular grid is used in the deep sea area, and the calculation is simplified by CGRID REG.

5. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: In (2), the modern physical process model adopts a nonlinear wave interaction model.

6. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: The interpolation in step 5 adopts linear interpolation: ， in, p(z) is the value of the air pressure at height z to be calculated. It can also be meteorological parameters such as temperature and wind speed. p bottom is the bottom z of a specific altitude layer bottom The corresponding variable value, p top is the top of a specific altitude layer z top The corresponding variable value, z, is the specific height that needs to be interpolated, between the two known heights.

7. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: The specific calculation of step 7 includes: The gradient Richardson number formula is as follows: ， Where g is the acceleration due to gravity; θ is the neutral temperature at the measurement height, in K, which can be derived from T, P, and PB in WRF, where T is the perturbed potential temperature, P is the perturbed air pressure, and PB is the ground state air pressure; u,v are the east-west and north-south wind speeds, in m / s; z is the height, obtained from PH and PHB of WRF, where PH is the disturbance potential and PHB is the ground potential; ∂θ / ∂ z and ∂u / ∂ z, ∂ v / ∂ z is the neutral temperature and the rate of change of the velocity component with height, which is calculated by the central difference approximation; The pulsating stress formula is as follows: ， ， in, PC_UW It is hoped that the u' 、 w' Mean Reynolds stress, in m 2 / s 2 , u' j,correlated yes j Point u The associated value of directional wind speed, and u Direction-independent wind speed u' j,independen Data is equal; v' j,correlated yes j Point v The associated value of the direction wind speed is obtained by u、v、w Direction-independent wind speed u' j,independen 、 v' j,independen and w' j,independen The data is combined and calculated; similarly, w' j,correlated yes j Point w The associated value of the direction wind speed is obtained by u' j,independen and w' j,independen The data is combined and calculated. a uv 、a vw 、a uw is the correlation coefficient, ensuring the correlation between different wind speed variables. U hub,correlated and W hub,correlated Refers to the corresponding correlation values ​​of the u component and w component of the velocity at the fan hub.

8. The method for simulating and analyzing a floating wind turbine in a wind and wave environment taking into account atmospheric stability according to claim 1, wherein: The OpenFAST simulation and analysis includes the following steps: Step 1: Set the wind file to be used; In OpenFAST's InflowWind INPUT FILE, set wind type to 3 = binary TurbSim FF; and add the path to the bts file in Parameters for Binary TurbSim Full-Field files; Step 2: Set the wave file to be used; Set WaveMod to 6 in OpenFAST's hydrodyn.dat and set the wave NetCDF file path in WvKinFile; Step 3: Run OpenFAST. Open a terminal in the file path and run OpenFAST; use MATLAB or PyDatView for subsequent data analysis.

Images