Meteorological ocean data prediction processing method and system

Abstract

The invention discloses a meteorological ocean data prediction processing method and system. The method comprises the following steps: acquiring position data of a geographic position for operation on the sea; acquiring meteorological data in a future preset time range in a space range corresponding to the position data; data related to sea waves and ocean currents are obtained from the meteorological data, and the related data are used for predicting the sea waves and the ocean currents; the sea waves and the ocean currents are predicted according to the related data, and sea wave data and ocean current data are obtained respectively; the presented meteorological data, the ocean wave data and the ocean current data are used as the basis for performing the operation in the spatial range and the preset time range; according to the method and the device, the problem of limited business guidance effect caused by only giving weather data when the business executed on the ocean is predicted in the prior art is solved, so that relatively reliable data support is provided for operation tasks on the ocean.

Description

Technical Field

[0001] This application relates to the field of marine meteorology. Specifically, it relates to a method and system for predicting and processing meteorological and oceanographic data. Background Art

[0002] Currently, there are a large number of operation tasks (which can also be called marine operations) on the ocean. Marine weather forecasts are very helpful for these ocean operations. For example, offshore wind power. With the advancement of energy transformation, offshore wind power, as one of the important sources of clean energy, is developing at an increasing speed. The development of offshore wind power depends on the weather. Therefore, the entities providing meteorological services will connect with major energy group enterprises to provide professional meteorological forecast and early warning data and manual services to help offshore wind farms reasonably utilize meteorological resources and avoid meteorological disaster risks in their own business scenarios.

[0003] In related technologies, numerical weather prediction models are usually used for numerical weather prediction in marine weather forecasting. First, a description of numerical weather prediction is given below.

[0004] Numerical weather prediction is a method of predicting the atmospheric motion state and weather phenomena in a future period by making numerical calculations on a large computer to solve the fluid dynamics and thermodynamics equations describing the weather evolution process under certain initial and boundary conditions based on the actual situation of the atmosphere.

[0005] Numerical weather prediction is different from the classical weather forecasting method based on synoptic meteorology. It is a quantitative and objective forecast. For this reason, numerical weather prediction requires the establishment of a better numerical prediction model reflecting the forecast period (for example, short-term, medium-term) and a calculation method with small errors, stable calculation, and relatively fast operation. Since numerical weather prediction needs to obtain meteorological data by various means (conventional observations, radar observations, ship observations, satellite observations, etc.), it is necessary to appropriately adjust, process, and objectively analyze meteorological data. Due to the extremely large amount of calculation data in numerical weather prediction, computers are usually used for it.

[0006] Based on the actual situation of the atmosphere, numerical weather prediction solves the fluid dynamics and thermodynamics equations describing the weather evolution process through numerical calculations under certain initial and boundary conditions, thereby obtaining the future weather conditions. Therefore, numerical weather prediction is different from the weather forecast made generally by the synoptic meteorology method combined with experience. This kind of forecast is a quantitative and objective forecast.

[0007] To improve the accuracy of numerical weather prediction, data assimilation is used to correct model errors and obtain initial conditions closer to the actual atmosphere. This technology combines observational data with the background field of a numerical model to generate a more accurate initial field. It can effectively integrate multi-source observational data (such as satellites, radars, ground stations, etc.) through mathematical methods (such as variational methods or Kalman filters) to correct model errors.

[0008] The inventors found that for the weather conditions of offshore wind farms, numerical weather prediction models often only provide weather conditions and do not incorporate real-time observational data. Such numerical prediction data is single, does not integrate ocean elements, and model errors are not corrected, thus failing to meet the requirements. Summary of the Invention

[0009] Embodiments of the present application provide a method and system for predicting and processing meteorological and ocean data to at least solve the problem of limited guidance for operations caused by only providing weather data when predicting operations performed on the ocean in related technologies.

[0010] According to one aspect of the present application, a method for predicting and processing meteorological and ocean data is provided. The method is applied in software, and the software is used to execute the method. The method includes the following steps: obtaining position data of a geographical location where operations are performed at sea; obtaining meteorological data within a future predetermined time range within the spatial range corresponding to the position data; obtaining data related to sea waves and ocean currents from the meteorological data, where the related data is used to predict the sea waves and the ocean currents; predicting the sea waves and the ocean currents based on the related data to obtain sea wave data and ocean current data respectively; presenting the meteorological data, the sea wave data, and the ocean current data, where the presented meteorological data, sea wave data, and ocean current data are used as a basis for performing operations within the spatial range and the predetermined time range.

[0011] Further, according to the spatial range and the predetermined time range configured by the user, where the spatial range is the geographical space within a predetermined range from the geographical location, and the predetermined time range is a time period; the time period is used to indicate that the meteorological data, the sea wave data, and the ocean current data are re-predicted at intervals of the time period.

[0012] Further, predict the power generation data based on the meteorological data, the sea wave data, and the ocean current data.

[0013] Further, input the meteorological data, the wave data, and the ocean current data into a pre-trained machine learning model. The machine learning model is a supervised machine learning model and is trained with multiple sets of training data. Each set of training data includes real meteorological data, wave data, ocean current data, and the corresponding power generation data of these data; obtain the predicted power generation data from the machine learning model.

[0014] Further, the meteorological data includes at least one of the following: temperature, humidity, air pressure, wind direction, wind speed; the wave data includes at least one of the following: significant wave height, mean wave period, wave direction; the ocean current data includes at least one of the following: water level, water flow velocity.

[0015] According to another aspect of the present application, there is also provided a meteorological and ocean data prediction and processing system, which is applied to software. The software includes the following modules: a first acquisition module for acquiring the location data of the geographical location where operations are carried out at sea; a second acquisition module for acquiring the meteorological data within a future predetermined time range within the spatial range corresponding to the location data; a third acquisition module for obtaining the data related to waves and ocean currents from the meteorological data, where the related data is used to predict the waves and the ocean currents; a first prediction module for predicting the waves and the ocean currents respectively to obtain wave data and ocean current data according to the related data; a presentation module for presenting the meteorological data, the wave data, and the ocean current data, where the presented meteorological data, wave data, and ocean current data are used as the basis for carrying out operations within the spatial range and the predetermined time range.

[0016] Further, the receiving module is used to receive the spatial range and the predetermined time range configured by the user. The spatial range is the geographical space within a predetermined range from the geographical location, and the predetermined time range is a time period; the time period is used to indicate that the meteorological data, the wave data, and the ocean current data are re-predicted every time interval of the time period.

[0017] Further, the second prediction module is used to predict the power generation data according to the meteorological data, the wave data, and the ocean current data.

[0018] Further, the second prediction module inputs the meteorological data, the wave data, and the ocean current data into a pre-trained machine learning model. The machine learning model is a supervised machine learning model and is obtained by training with multiple sets of training data. Each set of training data includes real meteorological data, wave data, ocean current data, and the power generation data corresponding to these data; and obtains the predicted power generation data from the machine learning model.

[0019] Further, the meteorological data includes at least one of the following: temperature, humidity, air pressure, wind direction, wind speed; the wave data includes at least one of the following: significant wave height, mean wave period, wave direction; the ocean current data includes at least one of the following: water level, water flow velocity.

[0020] In the embodiment of the present application, location data of a geographical location where operations are performed at sea is obtained; meteorological data within a future predetermined time range within the spatial range corresponding to the location data is obtained; data related to waves and ocean currents is obtained from the meteorological data, where the related data is used to predict the waves and the ocean currents; the waves and the ocean currents are predicted respectively based on the related data to obtain wave data and ocean current data; the meteorological data, the wave data, and the ocean current data are presented, and the presented meteorological data, wave data, and ocean current data are used as a basis for performing operations within the spatial range and the predetermined time range. By the present application, the problem that in the related art, only weather data is given when predicting operations performed on the ocean, resulting in limited guidance for operations, is solved, thereby providing relatively reliable data support for operation tasks on the ocean. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0022] Figure 1 is a schematic diagram of the WRF simulation area according to an embodiment of the present application;

[0023] Figure 2 is a schematic diagram of the FVCOM simulation area according to an embodiment of the present application;

[0024] Figure 3 is a schematic diagram of the WW3 simulation area according to an embodiment of the present application; and,

[0025] Figure 4 is a flowchart of the meteorological and ocean data prediction and processing method according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The following will detail this application with reference to the accompanying drawings and in combination with the embodiments.

[0027] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0028] In the following embodiments, numerical prediction simulations in the fields of meteorological science and ocean science are involved, including but not limited to numerically simulating and predicting the meteorological and ocean conditions of a predetermined area using specific algorithms and data. First, the technical terms involved in the following embodiments will be described.

[0029] Raster map

[0030] A raster map is essentially a form of discretization of the spatial and brightness characteristics of an image. It can be regarded as a matrix structure, where each element in the matrix, that is, a pixel, corresponds to a specific position in the image, and its value reflects the gray level of that point. Such an image can be analogized to a mosaic pattern composed of pixels. If all pixels have only two colors, black and white, then it is called a binary image, that is, a bitmap. If it contains multiple gray levels or colors, then it is a gray-scale image or a color image.

[0031] Spatio-temporal resolution

[0032] Spatio-temporal resolution is a geographical term that reflects the ability to capture the details of the changes of phenomena or things over time and space. It is usually the number of times a phenomenon or thing reappears per unit time and the actual area of the Earth's surface represented by a single raster cell.

[0033] Weather Research and Forecasting (WRF) Model

[0034] The WRF model is a mesoscale numerical weather prediction system designed specifically for atmospheric research and operational forecasting applications. It has two dynamic cores, a data assimilation system WRFDA (WRF Data Assimilation), and a software architecture that supports parallel computing and system scalability. The model is applicable to a wide range of meteorological applications with resolutions from dozens of meters to dozens of kilometers.

[0035] WRF can generate simulations based on actual atmospheric conditions (i.e., from observations and analyses) or idealized conditions. WRF provides a flexible and computationally efficient platform for operational forecasting, while reflecting the latest advances in physics, numerics, and data assimilation by developers from a wide research community. The WRF model is described below.

[0036] The WRF model can be initialized using idealized initialization or real data. An Eulerian mass solver, called the Advanced Research WRF Dynamics Solver, is supported in the model.

[0037] The WRF model also involves the WRF Preprocessing System (abbreviated as WPS). WPS is mainly used to define the simulation area and grid, interpolate the ground static data (such as terrain, land use type, soil parameters, etc.) obtained from external data sources into the simulation domain, and horizontally interpolate the meteorological data into the WRF simulation domain grid to provide the required data for the WRF main program.

[0038] The core of the WRF model is the main part for atmospheric simulation, which is divided into two cores: Advanced Research WRF (abbreviated as ARW) and Nonhydrostatic Mesoscale Model (abbreviated as NMM). The ARW (Advanced Research WRF) dynamics solver is the core component of the entire WRF model. It converts the data interpolated by the WPS preprocessing system into the initial field and boundary conditions in the specified format and performs the integration operation of the model.

[0039] The basic principle of WRF: The WRF model is written in Fortran 90 language and adopts a fully compressible and non-hydrostatic balance model. In the horizontal direction, it uses Arakawa C grid points, and in the vertical direction, it uses eta (terrain-following mass) coordinates. For time integration, it uses a third-order or fourth-order Runge-Kutta algorithm. The basic equations consist of the equations of motion, the continuity equation, the state equation, the thermodynamic equation, the water vapor equation, etc., which are obtained by introducing potential energy and potential temperature into the basic Navier-Stokes equations and performing terrain coordinate transformation.

[0040] WRF is based on finite difference or spectral discretization methods of these equations on a user-defined computational domain, which can be curvilinear or stretched, and the boundary conditions are set to be static, constant, or time-varying. This enables WRF to simulate a variety of weather and climate conditions, including convective and non-convective processes, precipitation, radiation, and surface processes. The output data of WRF simulations can take many forms, including gridded data of atmospheric variables such as air temperature, wind speed and direction, and precipitation, as well as vertical profiles of atmospheric variables. WRF outputs can be saved in a variety of file formats, compatible with other modeling and analysis tools, including the commonly used network data formats NetCDF and GRIB.

[0041] Input data for WRF: To initialize a WRF simulation, the user must provide initial conditions to define the atmospheric state at the start of the model run. These initial conditions may include atmospheric pressure, temperature, humidity, and wind speed and direction, as well as surface conditions such as soil moisture and vegetation cover. The WRF model can also obtain data from a variety of sources, including global and regional reanalysis data, satellite data, and observational data from weather stations and soundings. WRF can also obtain meteorological data at the model boundaries, called boundary conditions, to maintain consistency between the simulated atmospheric state and the atmospheric state outside the model domain. These data are typically obtained from global models.

[0042] The output data of WRF can take many forms, including gridded data of atmospheric variables such as air temperature, wind speed and direction, and precipitation, as well as vertical profiles of atmospheric variables. WRF outputs can be saved in a variety of file formats, compatible with other modeling and analysis tools, including the commonly used network data formats NetCDF and GRIB.

[0043] WRFDA is the data assimilation system of the WRF model, which is used to combine observational data with the model background field to generate a more accurate initial field and improve the accuracy of numerical weather prediction. It supports various algorithms such as three-dimensional variational 3DVAR, four-dimensional variational 4DVAR, and hybrid assimilation, is compatible with various observational data (such as surface stations, satellites, radars, etc.), and is widely used in weather forecasting, climate research, and environmental monitoring. WRFDA is open-source and efficient, and is an important tool for improving WRF simulations.

[0044] Ocean numerical model

[0045] The numerical prediction of ocean waves is carried out through wave models. Currently, the third-generation wave models include WAM (Wave Model), WAVE WATCH III, and SWAN (Simulating Waves Nearshore), etc. The main similarity among these models is the use of the dynamic spectrum balance equation including source and sink terms, but they differ in numerical methods and physical process parameterizations. WAM is built around a set of source terms, while SWAN and WW3 have a wide range of source terms and user-selectable parameterizations. WAM and WW3 were initially developed as large-scale models, while SWAN is specifically used to simulate nearshore ocean waves. Nevertheless, new source terms, parameterizations for nearshore processes, and more advanced numerical methods have also been introduced into WAM and WW3.

[0046] WAVE WATCH III (abbreviated as WW3) is a third-generation wave model developed by NCEP. The biggest difference from the WAM model lies in the calculation of the wind input function, and there are multiple schemes for the wind input function in the WW3 model. WW3 was initially only applicable to deep waters, and now the model has been corrected for shallow waters. The early versions of WW3 used spherical coordinates for horizontal grid points. This model not only considers the breaking of waves caused by shallow water, but also changes the coordinate system to an unstructured triangular grid or spherical multi-grid points.

[0047] Finite Volume Coastal Ocean Model

[0048] The Finite Volume Coastal Ocean Model (FVCOM for short) is a complex numerical model used to simulate ocean and coastal regions. Based on the finite volume method, it simulates the dynamics and transport processes of fluids by subdividing the water area into discrete finite volume cells. The FVCOM model has a wide range of applications in the fields of ocean science and ocean engineering and can accurately simulate the circulation characteristics of the ocean, including ocean currents, vortices, and boundary flows, etc. The model can also simulate the motion characteristics of water bodies, including tides, wind waves, and ocean waves, etc. In addition, the FVCOM model is also widely used in simulating water quality changes in the ocean and rivers, including water temperature, salinity, dissolved oxygen, and nutrients, etc.

[0049] In the following embodiments, considering the problems that the numerical model still has certain limitations, lacks ocean elements, does not combine meteorological and ocean buoy data of wind farms, and is difficult to meet the needs of meteorological and ocean forecasts, meteorological and ocean forecasts for a predetermined area are provided. For example, the forecast data may include: atmosphere, storm surge, and wave model.

[0050] In this embodiment, a method for predicting and processing meteorological and ocean data is provided. Figure 4 It is a flowchart of a method for predicting and processing meteorological and ocean data according to an embodiment of the present application, asFigure 4 As shown below, the steps involved in the method in Figure 4 will be described.

[0051] Step S402: Obtain the location data of the geographical location where the operation is carried out at sea.

[0052] Step S404: Obtain the meteorological data within a future predetermined time range within the spatial range corresponding to the location data.

[0053] In an additional embodiment, the operation type of the operation carried out at sea can be obtained, the data requirements corresponding to the operation type can be obtained according to the operation type, and the spatial range and the time range can be determined according to the data requirements; wherein, the time range and the spatial range input by the user when obtaining the meteorological data during the operation of the operation type in history are obtained, and the operation data is saved, and the operation data includes the time range and the spatial range, the operation type, and the operation time corresponding to the operation of the operation type; search for the corresponding spatial range and time range in the saved data according to the operation time and the operation type during the operation at sea.

[0054] There are many ways to obtain the meteorological data within the future predetermined time range. For example, obtain the current meteorological conditions, obtain the historical meteorological change law during the current time period according to the current time period, and determine the meteorological data within the future predetermined time range according to the current meteorological conditions and the meteorological change law; wherein, when there are multiple meteorological change laws during the historical time period, obtain multiple groups of meteorological conditions corresponding to each meteorological change law, mix the current meteorological conditions with all the obtained groups of meteorological conditions, classify the mixed meteorological conditions using machine learning, find the class where the current meteorological conditions are located, and find the meteorological change law corresponding to other groups of meteorological conditions in this class, and find the meteorological change law with the most occurrences in this class as the meteorological change law corresponding to the current meteorological conditions.

[0055] As an alternative embodiment, before obtaining the meteorological data within the future predetermined time range within the spatial range corresponding to the location data, the method further includes: receiving the spatial range and the predetermined time range configured by the user, wherein the spatial range is the geographical space within a predetermined range from the geographical location, and the predetermined time range is a time period; the time period is used to indicate that the meteorological data, the wave data, and the ocean current data are re-predicted at intervals of the time period.

[0056] Step S406: Obtain the data related to waves and ocean currents from the meteorological data, wherein the related data is used to predict the waves and the ocean currents.

[0057] Step S408, predict the ocean waves and the ocean current based on the relevant data to obtain ocean wave data and ocean current data respectively.

[0058] As an additional implementation manner, machine learning can be used to predict the ocean wave data and the ocean current data, which is described below. Obtain multiple sets of training data for training an ocean wave and ocean current model. Each set of training data is taken from real data in history. The training data includes weather data related to ocean waves and ocean currents, the time and location of weather data collection, and real ocean wave data and ocean current data. After the ocean current and ocean wave model training converges, input the relevant data, location data, and a future predetermined time range into the ocean wave and ocean current model, and obtain the predicted ocean wave data and ocean current data output from the ocean wave and ocean current model.

[0059] Step S410, present the meteorological data, the ocean wave data, and the ocean current data. The presented meteorological data, ocean wave data, and ocean current data are used as the basis for performing the operation within the spatial range and the predetermined time range.

[0060] The operation can be a power generation operation. The power generation here can be wind power generation and / or power generation using ocean waves and ocean currents, etc. At this time, the power generation data of the power generation can be predicted based on the meteorological data, the ocean wave data, and the ocean current data. There are many ways of prediction. For example, it can be performed by machine learning. In this example, input the meteorological data, the ocean wave data, and the ocean current data into a pre-trained machine learning model. The machine learning model is a supervised machine learning model, and the machine learning model is trained by multiple sets of training data. Each set of training data includes real meteorological data, ocean wave data, ocean current data, and the power generation data corresponding to these data. Obtain the predicted power generation data from the machine learning model.

[0061] The meteorological data in the above steps may include at least one of the following: temperature, humidity, air pressure, wind direction, wind speed; the ocean wave data may include at least one of the following: significant wave height, mean wave period, wave direction; the ocean current data may include at least one of the following: water level, water flow velocity.

[0062] By the above steps, the problem in the related art that the guidance for the operation is limited only by giving weather data when predicting the operations performed on the ocean is solved, thereby providing relatively reliable data support for the operation tasks on the ocean.

[0063] The following is an illustration with examples. In some of the following examples, data with higher spatial and temporal resolutions can also be provided. For example, the spatial resolution can be 1 km and the temporal resolution can be 15 minutes, so as to better meet the needs of maritime operations (such as wind power operations) in a predetermined area.

[0064] In the following embodiments, in order to provide refined meteorological and oceanographic data, three numerical models of meteorology, ocean waves, and ocean currents are used. Meteorological data can be generated by a meteorological system, such as by the WRF model. In one example, the meteorological data can include at least one of the following: temperature, humidity, air pressure, wind direction, and wind speed; ocean wave data can be generated by the WW3 model, and the ocean wave data can include at least one of the following: significant wave height, mean wave period, and wave direction; ocean current data can be generated by FVCOM (Finite Volume Community Ocean Model), and the ocean current data can include at least one of the following: water level and current velocity. A predetermined wind field can be selected as the wind field input for WW3 and FVCOM, thereby increasing the meteorological coupling in the ocean model and making the simulation results closer to the most realistic situation.

[0065] WRF has been widely used in meteorological numerical forecasting and simulation from the mesoscale to the global scale. In the following embodiments, the WRF model is used for local refined wind field forecasting. The spatial resolution of the model forecast product is 1 km, the temporal resolution is 15 minutes, and it is forecast once a day with a forecast period of up to 10 days. Figure 1 It is a schematic diagram of the WRF simulation area according to an embodiment of the present application, as Figure 1 shown. A double-layer nesting is adopted, and the projection method is the Lambert projection suitable for simulating the subtropical region. The central longitude and latitude are 121.1 and 34.65. The outer region d01 includes the eastern coastal areas of China, with a resolution of 5 km × 5 km and 140 × 200 grid cells. The horizontal resolution of the inner region is 1 km × 1 km, including the Jiangsu region d02 and the Shandong region d03, with 181 × 266 and 141 × 141 grid cells respectively. The model has 50 vertical layers, and the top of the model layer is 100 hPa. The 0.1-degree global numerical forecast data of the European Centre for Medium-Range Weather Forecasts (ECMWF) at 8:00 on the day before use is used as the initial field and boundary conditions of the regional model, and the boundary conditions are updated every 3 hours.

[0066] In order to obtain more accurate forecast results, this embodiment also uses data assimilation, which is a method of combining model forecasts with observation data to obtain the model initial field that is closest to the real atmosphere, thereby obtaining a more accurate forecast. In addition, this embodiment also uses the WRF supporting data assimilation system WRFDA, in which the observation data comes from the global telecommunication system GTS (Global Telecommunication System) and the wind farm SCADA (Supervisory Control And Data Acquisition) wind turbine data. Figure 1 The scattered points in the figure represent the horizontal distribution of observations used in a certain assimilation experiment, including ground airport, conventional observations, satellite wind, ship, sounding, and SCADA observations of wind farms. A three-dimensional variational method was used for data assimilation.

[0067] In an optional embodiment, in order to improve the forecast of boundary layer wind, this embodiment optimizes the WRF planetary boundary layer parameterization scheme and vertical stratification. Using Yonsei University (YSU) planetary boundary layer parameterization can reduce the error between forecast and observation and improve the correlation coefficient. In order to reflect the characteristics of near-surface layer wind, the vertical stratification is optimized, and the boundary layer is encrypted. The height of the first layer is adjusted from the default setting of 50 meters to 20 meters (dzbot=20 in the namelist.input configuration file), and the vertical stratification is adjusted (the low-level stretch coefficient dzstretch_s in the namelist.input configuration file is 1.1), and the model layer height below 200 meters is encrypted from four layers (0, 50.0, 113.9, 195.2) to eight layers (0, 20, 42, 66.2, 92.7, 122.0, 154.1, 189.4).

[0068] FVCOM uses the finite element method and dry-wet grids, which can not only well characterize the complex irregular coastlines and terrain geometry of estuaries and shelf areas, but also provide seawater backflow and flooding of land. The system uses unstructured arbitrary triangular grid units in the horizontal direction, which can locally encrypt key areas of concern and have the flexibility to fit the terrain. The grid feature allows the model calculation to fit the terrain, greatly improving the accuracy of the calculation. Physical quantities such as grid longitude and latitude, three-dimensional ocean currents, water level, water depth, and surface wind can be output on the unstructured grid.

[0069] Figure 2 is a schematic diagram of the FVCOM simulation area according to an embodiment of the present application, such as Figure 2As shown in the figure, in order to meet the high-precision requirements along the coast and balance the computational efficiency, the grid spacing is encrypted to 1 meter along the coast and 3 kilometers in the open sea. The output time resolution is 1 hour, and the forecasting period reaches 10 days. The wind field input is an important input data for the storm surge model, and the water level change caused by the wind can only be reflected through the wind. In the model, the wind field input file is required to be in the netCDF format, and this file is generated by the preprocessing script prepro.sh in the prepro directory. Mainly, the ten-meter U and V wind components are written into a specific netCDF file. Then, control FVCOM to read these elements during the calculation and participate in the calculation of the water level.

[0070] The WW3 model can simulate the wave field in the target sea area with high spatio-temporal resolution. The output wave parameters include significant wave height, mean wave period, peak frequency, peak direction, etc. The model spatial resolution is 1 kilometer, the time resolution is 1 hour, and the forecasting period is 10 days. Among them, the model topography data and water depth data come from the nautical chart, and the wind field driving data comes from ECMWF. Figure 3 It is a schematic diagram of the WW3 simulation area according to the embodiment of the present application, as Figure 3 shown. The longitude and latitude in the lower left corner are 119.6, 32.2, the resolution is 0.01 degree, and there are 241×281 grids, almost covering all coastal areas of the predetermined area.

[0071] Through the above implementation method, numerical forecast products with a time resolution of 15 minutes, a spatial resolution of 1 kilometer, and various meteorological and oceanographic elements can be output, enriching and improving the intelligent grid products (hourly time resolution) currently provided by the meteorological bureau. In addition, through historical meteorological and buoy observation data, machine learning correction of the forecast results can significantly improve the accuracy of the forecast results and enhance the usability of the forecast results.

[0072] The calculations of the three numerical models, WRF, FVCOM, and WW3, are illustrated with an example below.

[0073] The processes of WRF calculation and WRFDA data assimilation are as follows:

[0074] 1. Prepare the model and observation data. Download ECMWF coarse-grid meteorological forecast data, GTS, and SCADA observation data. After quality control of the SCADA wind turbine observation data, convert it into a format recognizable by WRFDA, and finally integrate it with the GTS data.

[0075] 2. Generate the model initial conditions and boundary conditions. Use the WRF preprocessing system WPS to interpolate the ECMWF coarse-grid data to the simulation area;

[0076] 3. Generate the assimilated background field. Integrate the initial and boundary conditions for 12 hours to obtain a 12-hour forecast, and use the forecast at this moment as the background field (Background) for data assimilation.

[0077] 4. Perform data assimilation. Use the WRFDA assimilation system and the three-dimensional variational assimilation algorithm (3D VAR) to assimilate the observations into the background field to generate the analysis field (Analysis).

[0078] 5. Use the above analysis field as the initial field of the model to continue integration and generate the final forecast result.

[0079] 6. Extract specific variables from the forecast results, including 10-meter wind, 10-meter gust, 2-meter air temperature, 2-meter relative humidity, visibility, precipitation rate, total irradiance, direct irradiance, scattered irradiance, etc.

[0080] 7. Horizontally interpolate the meteorological data according to the coordinates of the wind turbines, and then optimize the wind speed data using a machine learning model. The machine learning model is learned from historical wind speed observation data and forecast data and uses the LightGBM algorithm. The input elements include 10-meter wind speed, 10-meter wind direction, 2-meter air temperature, 2-meter humidity, and surface pressure, and the output is the corrected 10-meter wind.

[0081] The FVCOM calculation process is as follows:

[0082] 1. Read the driving field forecast data (here it represents the ECMWF 10-meter wind field);

[0083] 2. Prepare data such as grids and terrain. First, use the SMS software to create an unstructured grid of the simulation area; then, with the help of the fvcom-toolbox tool and the MATLAB TMD (Tidal Model Drive) global tidal prediction tool, read the created grid file to generate a series of files required by FVCOM, including grids, water depths, Coriolis forces, open boundaries, sponge boundaries, and tidal files.

[0084] 3. Execute FVCOM, specify the above file paths in the configuration file, specify the wind field file path, and turn on the wind field switch (WIND_ON is set to T).

[0085] 4. Interpolate the unstructured grid data onto a structured grid for easy visualization by developers.

[0086] 5. Extract the storm surge height data at the specified location and post-process and optimize the results.

[0087] The WW3 calculation process is as follows:

[0088] 1. Use the gridgen tool to create WW3 grids and data files. The creation method refers to examplels / create_grid.m to generate depth, mask, and grid files.

[0089] 2. Modify the initial field of the ww3_strt.inp configuration mode.

[0090] 3. Modify the wind field data input in the ww3_prnc.inp configuration.

[0091] 4. Modify the multiple nesting mode in the ww3_multi.inp configuration.

[0092] 5. Execute WW3 to output wave field files, including significant wave height, wave direction, mean wave period, maximum wave height, wind wave, and swell wave.

[0093] 6. Interpolate the wave point file to the buoy site and then process it. For example, machine learning algorithms can be used for post-processing optimization.

[0094] In this embodiment, an electronic device is provided, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the method in the above embodiment.

[0095] The above program can run in the processor or can also be stored in the memory (or referred to as a computer-readable medium). The computer-readable medium includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassette tapes, magnetic disk storage, or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0096] These computer programs can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate computer-implemented processing. Thus, the instructions executed on the computer or other programmable device provide for implementation in the process Figure 1 one process or multiple processes and / or blocks Figure 1Steps for the functions specified in one or more boxes can be implemented by different modules corresponding to different steps.

[0097] This embodiment provides such a device or system. The system is called a meteorological and ocean data prediction and processing system and is applied in software. The software includes the following modules: a first acquisition module for acquiring position data of a geographical location where operations are carried out at sea; a second acquisition module for acquiring meteorological data within a future predetermined time range within the spatial range corresponding to the position data; a third acquisition module for acquiring data related to sea waves and ocean currents from the meteorological data, where the related data is used to predict the sea waves and the ocean currents; a first prediction module for predicting the sea waves and the ocean currents respectively to obtain sea wave data and ocean current data according to the related data; a presentation module for presenting the meteorological data, the sea wave data and the ocean current data, where the presented meteorological data, sea wave data and ocean current data are used as a basis for carrying out operations within the spatial range and the predetermined time range.

[0098] Optionally, the receiving module is used to receive the spatial range and the predetermined time range configured by the user, where the spatial range is a geographical space within a predetermined range from the geographical location, and the predetermined time range is a time period; the time period is used to indicate that the meteorological data, the sea wave data and the ocean current data are re-predicted every time interval of the time period.

[0099] Optionally, the second prediction module is used to: predict the power generation data according to the meteorological data, the sea wave data and the ocean current data.

[0100] Optionally, the second prediction module is used to: input the meteorological data, the sea wave data and the ocean current data into a pre-trained machine learning model, where the machine learning model is a supervised machine learning model, and the machine learning model is obtained by training with multiple sets of training data. Each set of training data includes real meteorological data, sea wave data and ocean current data, as well as the power generation data corresponding to these data; obtain the predicted power generation data from the machine learning model.

[0101] As another additional implementation manner, the meteorological data includes at least one of the following: temperature, humidity, air pressure, wind direction, wind speed; the sea wave data includes at least one of the following: significant wave height, mean wave period, wave direction; the ocean current data includes at least one of the following: water level, water flow velocity.

[0102] The system or device is used to implement the functions of the methods in the above embodiments. Each module in the system or device corresponds to each step in the method, and those that have been described in the method will not be elaborated here.

[0103] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for forecasting and processing meteorological and oceanographic data, characterized in that: include: Obtaining location data of geographical locations where operations are conducted offshore; Acquire meteorological data within a future predetermined time range within a spatial range corresponding to the location data; Acquire data related to ocean waves and ocean currents from the meteorological data, wherein the related data is used to predict the ocean waves and ocean currents; Predicting the ocean waves and ocean currents according to the relevant data to obtain ocean wave data and ocean current data respectively; The meteorological data, the ocean wave data and the ocean current data are presented, wherein the presented meteorological data, the ocean wave data and the ocean current data are used as a basis for performing the operation within the spatial range and the predetermined time range.

2. The method for forecasting and processing meteorological and oceanographic data according to claim 1, characterized in that: Before obtaining the meteorological data within a future predetermined time range within the spatial range corresponding to the location data, the method further includes: Receive the spatial range and the predetermined time range configured by the user, wherein the spatial range is the geographical space within the predetermined range from the geographical location, and the predetermined time range is a time period; the time period is used to indicate that the meteorological data, the wave data and the current data are re-forecasted at intervals of the time period.

3. The method for forecasting and processing meteorological and oceanographic data according to claim 1, characterized in that: The operation performed at sea is power generation, and the method further comprises: The power generation data of the power generation is predicted based on the meteorological data, the ocean wave data and the ocean current data.

4. The method for forecasting and processing meteorological and oceanographic data according to claim 3, characterized in that: Predicting the power generation data of the wind power generation includes: Inputting the meteorological data, the ocean wave data and the ocean current data into a pre-trained machine learning model, wherein the machine learning model is a supervised machine learning model, and the machine learning model is trained by multiple sets of training data, each set of training data includes real meteorological data, ocean wave data and ocean current data and power generation data corresponding to these data; Obtain predicted power generation data from the machine learning model.

5. The method for forecasting and processing meteorological and oceanographic data according to any one of claims 1 to 4, characterized in that: The meteorological data includes at least one of the following: temperature, humidity, air pressure, wind direction, and wind speed; the ocean wave data includes at least one of the following: effective wave height, average wave period, and wave direction; the ocean current data includes at least one of the following: water level and water flow speed.

6. A meteorological and oceanographic data prediction and processing system, characterized in that: include: A first acquisition module is used to acquire location data of a geographical location where operations are performed at sea; A second acquisition module is used to acquire meteorological data within a future predetermined time range within a spatial range corresponding to the location data; A third acquisition module is used to acquire data related to ocean waves and ocean currents from the meteorological data, wherein the related data is used to predict the ocean waves and ocean currents; A first prediction module, used for predicting the ocean waves and ocean currents according to the relevant data to obtain ocean wave data and ocean current data respectively; A presentation module is used to present the meteorological data, the ocean wave data and the ocean current data, wherein the presented meteorological data, the ocean wave data and the ocean current data are used as a basis for performing the operation within the spatial range and the predetermined time range.

7. The meteorological and oceanographic data prediction and processing system according to claim 6, characterized in that: The system further comprises: A receiving module is used to receive the spatial range and the predetermined time range configured by the user, wherein the spatial range is the geographical space within the predetermined range from the geographical location, and the predetermined time range is a time period; the time period is used to indicate that the meteorological data, the wave data and the current data are re-forecasted at intervals of the time period.

8. The meteorological and oceanographic data prediction and processing system according to claim 6, characterized in that: The operation performed at sea is power generation, and the system further comprises: The second prediction module is used to predict the power generation data according to the meteorological data, the wave data and the current data.

9. The meteorological and oceanographic data prediction and processing system according to claim 8, characterized in that: The second prediction module is used for: Inputting the meteorological data, the ocean wave data and the ocean current data into a pre-trained machine learning model, wherein the machine learning model is a supervised machine learning model, and the machine learning model is trained by multiple sets of training data, each set of training data includes real meteorological data, ocean wave data and ocean current data and power generation data corresponding to these data; Obtain predicted power generation data from the machine learning model.

10. The meteorological and oceanographic data prediction and processing system according to any one of claims 6 to 9, characterized in that: The meteorological data includes at least one of the following: temperature, humidity, air pressure, wind direction, and wind speed; the ocean wave data includes at least one of the following: effective wave height, average wave period, and wave direction; the ocean current data includes at least one of the following: water level and water flow speed.

Images