Space-borne laser radar wind field UV wind vector inversion method, device and equipment

Abstract

The embodiment of the invention provides a spaceborne laser radar wind field UV wind vector inversion method, device and equipment, and is applied to the technical field of atmospheric detection. The method comprises the following steps: acquiring optimal pairing data of a satellite horizontal sight wind speed, a data profile top height, a data profile bottom height, a site latitude and longitude, a satellite azimuth angle and a sounding wind speed and wind direction of a satellite-borne laser radar wind field, inputting the optimal pairing data into a constructed wind speed and wind direction inversion algorithm model, and outputting a UV wind vector. In this way, a satellite-borne laser Doppler radar wind field inversion model, namely a wind speed and wind direction inversion algorithm model, is established based on ERA5 analysis field data, Beidou sounding data, foundation wind profile radar data and satellite data by using a plurality of machine learning methods based on a satellite-borne laser Doppler radar wind field inversion technology of machine learning; the function of obtaining the UV wind vector, namely the weft wind component and the warp wind component, through horizontal sight line wind observation inversion is successfully achieved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, in particular to the field of atmospheric detection technology, and specifically to an inversion method, device and equipment for UV wind vector in a wind field of a satellite-borne laser radar. Background Art

[0002] Wind fields are a crucial meteorological element in the study of atmospheric dynamics and climate change, and are also a key observable variable for atmospheric sounding. To directly measure global wind profiles, the Aeolus satellite, equipped with a laser Doppler wind radar, was recently launched. This satellite successfully achieved the first direct observation of Earth's wind field. However, the remote sensing observations obtained by the Aeolus satellite only capture the component of Earth's atmospheric wind field along the line of sight between the satellite and the target, and these observations are subject to certain errors and biases, making them unusable directly. Summary of the Invention

[0003] The present disclosure provides a method, apparatus, device and storage medium for inverting UV wind vectors in a spaceborne lidar wind field.

[0004] According to a first aspect of the present disclosure, a method for inverting UV wind vectors in a spaceborne lidar wind field is provided. The method comprises:

[0005] Obtaining optimal paired data for the satellite-borne lidar wind field; the optimal paired data includes the satellite horizontal line-of-sight wind speed, the height of the top of the data profile, the height of the bottom of the data profile, the longitude and latitude of the station, the satellite azimuth, and the wind speed and direction of the sounding; the optimal paired data is determined based on satellite data, ERA5 data, Beidou sounding data, and ground-based wind profiler radar data;

[0006] The optimal paired data is input into the constructed wind speed and direction inversion algorithm model to output the UV wind vector. The wind speed and direction inversion algorithm model uses a support vector machine regression algorithm and a min-max normalization model to perform data inversion.

[0007] According to the above aspects and any possible implementation, a further implementation is provided, wherein the construction of the wind speed and direction inversion algorithm model includes:

[0008] Obtaining historical optimal paired data, the historical optimal paired data including historical satellite horizontal line-of-sight wind speed, historical data profile top height, historical data profile bottom height, historical station latitude and longitude, historical satellite azimuth, historical sounding wind speed and direction, and historical ERA5 UV wind vector;

[0009] Preprocessing the historical optimal pairing data based on a preset min-max normalization model;

[0010] Based on the support vector machine regression algorithm, using the preprocessed historical satellite horizontal line-of-sight wind speed, historical data profile top height, historical data profile bottom height, historical station longitude and latitude, historical satellite azimuth angle, and historical radiosonde wind speed and direction as samples, and using the historical ERA5 UV wind vector as the annotation, the obtained training set is used to train a preset Fengshen wind speed and direction inversion algorithm model to obtain a constructed wind speed and direction inversion algorithm model;

[0011] Among them, the training of the preset Fengshen wind speed and direction inversion algorithm model includes: finding the optimal parameters of the preset Fengshen wind speed and direction inversion algorithm model based on grid search and cross-validation, and using the optimal parameters to train the preset Fengshen wind speed and direction inversion algorithm model.

[0012] In the above-mentioned aspects and any possible implementation manners, a further implementation manner is provided. The obtaining of the historical optimal paired data includes:

[0013] Obtain the satellite matchable data set and the ERA5 and ground-based data match data set;

[0014] According to the preset correspondence relationship between the satellite - ERA5 and the ground-based data height layers, convert the wind speed and direction in the multi-layer ERA5 and ground-based data match data corresponding to the satellite matchable data set height layer into UV wind vectors;

[0015] Perform arithmetic averaging on the UV wind vectors to obtain the mean UV wind vector;

[0016] Convert the mean UV wind vector into the mean wind speed and direction;

[0017] Based on the preset satellite wind speed conversion algorithm, calculate the satellite horizontal line-of-sight wind speed according to the mean wind speed and direction;

[0018] According to the satellite horizontal line-of-sight wind speed, match the satellite matchable data set and the ERA5 and ground-based data match data set to obtain the historical optimal paired data.

[0019] In the above-mentioned aspects and any possible implementation manners, a further implementation manner is provided. Obtaining the ERA5 and ground-based data match data set includes:

[0020] Obtain ERA5 data and ground-based data, and the ground-based data includes Beidou radiosonde data and ground-based wind profile radar data;

[0021] According to the preset screening rules for the matchable data between Beidou radiosonde and ERA5, obtain the matchable dataset of ERA5 and Beidou radiosonde data; wherein, the preset screening rules for the matchable data between Beidou radiosonde and ERA5 include: the ERA5 data meets the preset spatial distance requirement from the ground-based station, and the Beidou radiosonde data and the ERA5 data are within the first preset nearest time and the preset air pressure difference.

[0022] According to the preset screening rules for the matchable data between ground-based wind profile and ERA5, obtain the matchable dataset of ERA5 and ground-based data; wherein, the preset screening rules for the matchable data between ground-based wind profile and ERA5 include: the matchable dataset of ERA5 and Beidou radiosonde data and the ERA5 data are within the second preset nearest time and the preset height difference.

[0023] For the aspects and any possible implementation manners as described above, further provide an implementation manner, the obtaining of the ERA5 data and the ground-based data includes:

[0024] When the ERA5 data meets the preset spatial distance requirement from the ground-based station, obtain the corresponding ERA5 data.

[0025] Based on the ERA5 data, obtain the Beidou radiosonde data and the ground-based wind profiler data within the preset time.

[0026] For the aspects and any possible implementation manners as described above, further provide an implementation manner, the obtaining of the satellite matchable dataset includes:

[0027] According to the preset screening rules for the matchable satellite data, obtain the satellite matchable dataset.

[0028] Wherein, the preset screening rules for the matchable satellite data include: the time of the satellite wind profile data is within the preset matching time period, the longitude and latitude of the satellite wind profile data are within the preset longitude and latitude range of the ground-based station, and the satellite passes over the ground-based station.

[0029] For the aspects and any possible implementation manners as described above, further provide an implementation manner, the method further includes:

[0030] Convert the UV wind vector into wind speed and wind direction.

[0031] Based on the preset evaluation indexes and the preset evaluation dimensions, evaluate the wind speed and wind direction respectively with the wind profile data, the ERA� data and the Beidou radiosonde data; the preset evaluation indexes include the correlation coefficient, the root mean square error, the mean absolute error and the mean deviation, and the preset evaluation dimensions include the station dimension, the height layer dimension, the month dimension and the integrity dimension.

[0032] According to a second aspect of the present disclosure, there is provided an apparatus for retrieving the UV wind vector of the on-board lidar wind field. The apparatus includes:

[0033] An acquisition module, configured to acquire the optimal paired data of the on-board lidar wind field; the optimal paired data includes the satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, the site longitude and latitude, the satellite azimuth angle, and the wind speed and direction of the radiosonde; the optimal paired data is determined according to satellite data, ERA5 data, Beidou radiosonde data, and ground-based wind profile radar data;

[0034] An inversion module, configured to input the optimal paired data into a pre-constructed wind speed and direction inversion algorithm model, and output the UV wind vector. The wind speed and direction inversion algorithm model uses the support vector machine regression algorithm and the min-max normalization model for data inversion.

[0035] According to a third aspect of the present disclosure, there is provided an electronic device. The electronic device includes: a memory and a processor. A computer program is stored on the memory, and when the processor executes the program, the method described above is implemented.

[0036] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0037] The method, apparatus, device, and storage medium for retrieving the UV wind vector of the on-board lidar wind field provided by the embodiments of the present application can acquire the optimal paired data of the on-board lidar wind field, including the satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, the site longitude and latitude, the satellite azimuth angle, and the wind speed and direction of the radiosonde, and input the optimal paired data into a pre-constructed wind speed and direction inversion algorithm model to output the UV wind vector. Based on this, the on-board lidar Doppler radar wind field inversion technology based on machine learning uses a variety of machine learning methods, and based on ERA5 analysis field data, Beidou radiosonde data, wind profile radar data, and satellite data, an on-board lidar Doppler radar wind field inversion model, that is, a wind speed and direction inversion algorithm model, is established, and the function of retrieving the UV wind vector, that is, the zonal wind and meridional wind components, from the horizontal line-of-sight wind observation is successfully realized.

[0038] It should be understood that the content described in the summary of the invention section is not intended to limit the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. Brief Description of the Drawings

[0039] In conjunction with the accompanying drawings and with reference to the following detailed description, the above and other features, advantages, and aspects of the embodiments of the present disclosure will become more apparent. The drawings are used to better understand the solution and do not limit the present disclosure. In the drawings, the same or similar reference numerals denote the same or similar elements, where:

[0040] Figure 1 A flowchart of a method for inverting the UV wind vector of the spaceborne lidar wind field according to an embodiment of the present disclosure is shown;

[0041] Figure 2 A block diagram of a device for inverting the UV wind vector of the spaceborne lidar wind field according to an embodiment of the present disclosure is shown;

[0042] Figure 3 A block diagram of an exemplary electronic device capable of implementing the embodiments of the present disclosure is shown. Detailed implementation manners

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Apparently, the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure.

[0044] In addition, the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the front and rear associated objects.

[0045] In the present disclosure, a spaceborne lidar Doppler radar wind field inversion technology based on machine learning uses various machine learning methods to establish a spaceborne lidar Doppler radar wind field inversion model, that is, a wind speed and wind direction inversion algorithm model, based on ERA5 analysis field data, Beidou sounding data, wind profiler radar data, and satellite data, and successfully realizes the function of inverting the UV wind vector, that is, the zonal wind and meridional wind components, from the horizontal line-of-sight wind observation.

[0046] Figure 1 A flowchart of a method 100 for inverting the UV wind vector of the spaceborne lidar wind field according to an embodiment of the present disclosure is shown.

[0047] At block 110, obtain the optimal paired data of the spaceborne lidar wind field; the optimal paired data includes the satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, the site longitude and latitude, the satellite azimuth angle, and the radiosonde wind speed and direction; the optimal paired data is determined based on satellite data, ERA5 data, Beidou radiosonde data, and ground-based wind profile radar data.

[0048] In some embodiments, the optimal paired data can be screened from the original paired data, and the effective samples of Rayleigh-clear and Mie-cloudy in the satellite are retained, that is, the samples in the state where the scattering phenomenon in the atmosphere is weak and the line of sight is relatively clear are retained, forming the optimal paired data. Among them, the satellite horizontal line-of-sight (HLOS) wind speed, the top height of the data profile (the top height of the Bin), the bottom height of the data profile (the bottom height of the Bin), the site longitude and latitude, the satellite azimuth angle, and the radiosonde wind speed and direction will be used as the input of the wind speed and direction inversion algorithm model.

[0049] In some embodiments, the original paired data can be a data set determined by one-by-one matching of satellite data, ERA5 data, Beidou radiosonde data, and ground-based wind profile radar data.

[0050] At block 120, input the optimal paired data into the constructed wind speed and direction inversion algorithm model, and output the UV wind vector. The wind speed and direction inversion algorithm model uses the support vector machine regression algorithm and the min-max normalization model for data inversion.

[0051] In some embodiments, the UV wind vector can be the UV wind vector of ERA5. In the UV wind vector, U and V respectively represent the components of the wind in the east-west direction and the north-south direction, that is, the zonal wind and meridional wind components.

[0052] In some embodiments, the construction of the above wind speed and direction inversion algorithm model includes:

[0053] Obtain historical optimal paired data, which includes historical satellite horizontal line-of-sight wind speed, historical top height of the data profile, historical bottom height of the data profile, historical site longitude and latitude, historical satellite azimuth angle, historical radiosonde wind speed and direction, and historical ERA5 UV wind vector;

[0054] Based on the preset min-max normalization model, preprocess the historical optimal paired data;

[0055] Based on the Support Vector Regression (SVR) algorithm, using the preprocessed historical satellite horizontal line-of-sight wind speed, historical data profile top height, historical data profile bottom height, historical site longitude and latitude, historical satellite azimuth angle, and historical radiosonde wind speed and direction as samples, and the historical ERA5 UV wind vector as the annotation, the obtained training set is used to train a preset Aeolus wind speed and direction inversion algorithm model to obtain a constructed wind speed and direction inversion algorithm model;

[0056] Among them, training the preset Aeolus wind speed and direction inversion algorithm model includes: finding the optimal parameters of the preset Aeolus wind speed and direction inversion algorithm model based on grid search and cross-validation, and using the optimal parameters to train the preset Aeolus wind speed and direction inversion algorithm model.

[0057] In some embodiments, first, min-max normalization is used for data preprocessing. This process is mainly a linear transformation of the historical optimal paired data. By subtracting the data from the minimum value data and then dividing by the difference between the maximum and minimum values, the data is mapped between 0 and 1. Then, grid search and cross-validation are used to find the optimal parameters of the preset Aeolus wind speed and direction inversion algorithm model, and the finally confirmed parameters are used to train and save the SVR model and the min-max normalization model. Among them, grid search includes: given a search range, and then determining the optimal value by searching all points within the search range; cross-validation includes: randomly dividing the original data into N parts, each time selecting N - 1 parts as the training set, and the remaining one part as the validation set. A model is obtained using the training set, and then the validation set is used for testing to obtain evaluation metrics. This step is repeated N times, and the average value of the evaluation metrics after these N experiments is obtained to get the accuracy of this model.

[0058] In some embodiments, the above-mentioned obtaining of historical optimal paired data includes:

[0059] Obtaining a satellite matchable dataset and an ERA5 and ground-based data match dataset;

[0060] According to the preset corresponding relationship between the satellite - ERA5 and ground-based data height levels, the wind speed and direction in the multi-layer ERA5 and ground-based data match data corresponding to the height level of the satellite matchable dataset are converted into UV wind vectors;

[0061] Performing arithmetic averaging on the UV wind vectors to obtain the mean UV wind vector;

[0062] Converting the mean UV wind vector into the average wind speed and direction;

[0063] Based on the preset satellite wind speed conversion algorithm, calculating the satellite horizontal line-of-sight wind speed according to the average wind speed and direction;

[0064] Match the satellite-matched dataset and the ERA5 and ground-based data-matched dataset according to the satellite horizontal line-of-sight wind speed to obtain the historical optimal paired data.

[0065] In some embodiments, since multiple ground-based wind profile data (ground-air data) height levels correspond to one satellite wind profile data (Aeolus data) height level, within the same height level range, the ground-based radar can observe 100 data values, while the spaceborne radar can only observe 1 data value. Therefore, the multi-layer ground-air data needs to be processed into data consistent with the Aeolus data height level. Specifically, first, convert the wind speed and direction of the multi-layer ground-air data into UV wind vectors, then perform arithmetic averaging respectively, and convert the mean value of the multi-layer ground-air UV wind vectors back into wind speed and direction. Since the Aeolus satellite measures the horizontal line-of-sight (HLOS) wind speed of the atmosphere, the wind speed and direction of the multi-layer averaged ground-air data need to be converted into the HLOS wind speed of the Aeolus satellite.

[0066] In some embodiments, the preset correspondence between the satellite-ERA5 and ground-based data height levels is determined according to the bottom height of the data profile, the top height of the data profile, and the data height.

[0067] In some embodiments, the specific process of obtaining the historical optimal paired data includes:

[0068] Step 18: Determine that multiple ERA5 and ground-based data-matched data height levels correspond to one Aeolus data height level;

[0069] Step 19: Perform arithmetic averaging on the wind speed and direction of the ERA5 and ground-based data-matched data for each layer;

[0070] Step 20: Convert the multi-layer averaged wind speed and direction into HLOS wind speed (conversion formula: cos(wind speed line-of-sight angle of Aeolus - ERA5 wind direction) * ERA5 wind speed); form the ERA5 and Aeolus matched dataset;

[0071] Step 21: Merge the datasets of ERA5, Beidou radiosonde, ground-based wind profile, and Aeolus that are mutually matched into one file to form the historical optimal paired data.

[0072] Among them, the conversion formula includes:

[0073]

[0074] Among them, represents the ground-based horizontal wind speed converted to the atmospheric horizontal line-of-sight direction, represents the direction angle of the satellite data (Aeolus), that is, the data in "wind_result_los_azimuth", d tThe horizontal wind direction indicating the wind direction in the ground-based data, s t Indicates the magnitude of the ground-based horizontal wind speed.

[0075] In summary, the process of obtaining the historical optimal paired data is the matching process of ERA5 data and the Fengshen dataset, for the inversion of the UV wind vector of the spaceborne lidar wind field and the construction of the wind speed and wind direction inversion algorithm model.

[0076] In some embodiments, obtaining the above-mentioned matching dataset of ERA5 and ground-based data includes:

[0077] Obtain ERA5 data and ground-based data, where the ground-based data includes Beidou sounding data and ground-based wind profiler radar data;

[0078] According to the preset screening rules for the matchable data between Beidou sounding and ERA5, obtain the matching dataset of ERA5 and Beidou sounding data; among them, the preset screening rules for the matchable data between Beidou sounding and ERA5 include: the ERA5 data meets the preset spatial distance requirement for the ground-based station, and the Beidou sounding data and the ERA5 data are within the first preset nearest time and the preset pressure difference.

[0079] According to the preset screening rules for the matchable data between the ground-based wind profiler and ERA5, obtain the matching dataset of ERA5 and ground-based data; among them, the preset screening rules for the matchable data between the ground-based wind profiler and ERA5 include: the matching dataset of ERA5 and Beidou sounding data and the ERA5 data are within the second preset nearest time and the preset height difference.

[0080] In some embodiments, the preset screening rules for the matchable data between Beidou sounding and ERA5, the preset screening rules for the matchable data between the ground-based wind profiler and ERA5, the first preset nearest time, the second preset nearest time, the preset height difference, and the preset pressure difference can be set according to the user's time requirements. Among them, the first preset nearest time and the second preset nearest time can be set to the same time.

[0081] In some embodiments, the specific process of obtaining the matching dataset of ERA5 and ground-based data includes:

[0082] Step 16. For the matching of ERA5 data and Beidou radiosonde data, that is, the acquisition of the ERA5 and Beidou radiosonde data matching dataset. First, perform a spatial matching judgment on the 0.25° * 0.25° grid data of ERA5 and the ground-based stations, that is, whether the wind profile data of ERA5 is within 100 km of the longitude and latitude of the current ground-based station; if the spatial distance between the two exceeds 100 km, end the current judgment; if the spatial matching is successful, find the nearest time data of the Beidou radiosonde data and the ERA5 wind profile data. Subsequently, convert the UV wind vector of ERA5 into wind speed and direction, and use the Beidou radiosonde data with the smallest difference in air pressure for each layer of wind speed and direction of ERA5. Based on the fact that ERA5 has no altitude, use the matched Beidou radiosonde altitude data as the altitude of ERA5 to form the ERA5 and Beidou radiosonde data matching dataset;

[0083] Step 17. For the matching of ERA5 data and ground-based wind profiler radar data, that is, the acquisition of the ERA5 and ground-based data matching dataset. First, find the nearest time data of the ground-based wind profiler radar data and the ERA5 data in the ERA5 and ground-based data matching dataset. Subsequently, use the ground-based wind profiler radar data with the smallest difference in height for each layer of wind speed and direction of ERA5 to form the ERA5 and ground-based data matching dataset.

[0084] As can be seen from the above, the process of obtaining the ERA5 and ground-based data matching dataset is the matching process of ERA5 data and ground-based data, for the inversion of the UV wind vector of the spaceborne lidar wind field and the construction of the wind speed and direction inversion algorithm model.

[0085] In some embodiments, the above acquisition of ERA5 data and ground-based data includes:

[0086] When the ERA5 data meets the preset spatial distance requirement from the ground-based station, obtain the corresponding ERA5 data;

[0087] Based on the ERA5 data, obtain the Beidou radiosonde data and ground-based wind profiler radar data within the preset time.

[0088] In some embodiments, the specific process of obtaining ERA5 data and ground-based data includes:

[0089] Step 12. Determine that the data source of the ground-based data is obtained from the meteorological big data cloud platform ("Tianqing"); the data source of the ERA5 data is from the official website of the European Centre for Medium-Range Weather Forecasts (ECMWF);

[0090] Step 13: For the acquisition of ERA5 data, the data within the entire territory can be automatically downloaded through the official website script. When the ERA5 data meets the preset spatial distance requirement from the ground-based stations, the corresponding ERA5 data (NC format) is acquired. The ERA5 data includes: data longitude: "longitude"; data latitude: "latitude"; data generation time: "valid_time"; pressure level: "pressure_level"; meridional wind: "U"; zonal wind: "V", etc.

[0091] Step 14: For the acquisition of Beidou radiosonde data, based on the ERA5 data, the Beidou radiosonde data within the specified time range can be downloaded from Tianqing first and quality-controlled. The Beidou radiosonde data includes wind direction, wind speed, data altitude, pressure, quality control information (quality control code), etc. Among them, the specified time range can be the range matching the time of the ERA5 data.

[0092] Step 15: For the acquisition of ground-based wind profiler radar data, based on the ERA5 data, the ground-based wind profiler radar data within the specified time range can be downloaded from Tianqing first and quality-controlled. The ground-based wind profiler radar data includes wind direction, wind speed, data altitude, pressure, quality control information (quality control code), etc. Among them, the specified time range can be the range matching the time of the ERA5 data.

[0093] In summary, the process of acquiring ERA5 data and ground-based data is the preparation process for ERA5 data and ground-based data, in order to prepare for the inversion of the UV wind vector of the spaceborne lidar wind field and the construction of the wind speed and wind direction inversion algorithm model.

[0094] In some embodiments, the acquisition of the above satellite matchable dataset includes:

[0095] According to the preset satellite matchable data screening rules, the satellite matchable dataset is acquired;

[0096] Among them, the preset satellite matchable data screening rules include: the time of the satellite wind profile data is within the preset matching time period, the longitude and latitude of the data profile of the satellite wind profile data are within the preset ground-based station longitude and latitude range, and the satellite passes over the ground-based station.

[0097] In some embodiments, the satellite wind profile data can be determined based on the original observation dataset of the "Aeolus" meteorological satellite. Among them, the original observation dataset of the "Aeolus" meteorological satellite can be based on the automatic download of Aeolus satellite data from the official website of the European Space Agency (ESA) through a script, that is, the wind profile data of each station.

[0098] In some embodiments, the preset satellite match data screening rules, preset matching time period, and preset ground station longitude and latitude range can be set according to the actual needs of the user. For example, the preset ground station longitude and latitude range can be set within 100 km of the longitude and latitude of the ground-based radar station that matches the satellite.

[0099] In some embodiments, the specific process of obtaining the satellite match data set includes:

[0100] Step 1: Determine the time period to be matched, that is, determine the preset matching time period;

[0101] Step 2: Determine that the satellite wind profile data source is to obtain Aeolus satellite data from the ESA official website;

[0102] Step 3: Download the profile data of each ground station through the official website script;

[0103] Step 4: Determine whether the Aeolus satellite passes over the current ground station;

[0104] Step 5: If the Aeolus satellite passes over the current ground station, perform a spatial matching judgment, that is, whether the Aeolus satellite wind profile data is the satellite wind profile data within 100 km of the longitude and latitude of the current ground station;

[0105] Step 6: If the Aeolus satellite does not pass over the current ground station, end the current judgment and return to Step 3 to continue obtaining the satellite match data set for the next station;

[0106] Step 7: If the spatial matching is successful, download the Aeolus data (NC format) through the official website script;

[0107] Step 8: If the spatial matching is unsuccessful, end the current judgment and return to Step 3 to continue obtaining the satellite match data set for the next station;

[0108] Step 9: Parse the Aeolus data and determine whether the parsing is successful;

[0109] Step 10: If the Aeolus data is successfully parsed, a satellite matchable data set is formed (the satellite matchable data set belongs to the original observation data set of the Aeolus meteorological satellite); the satellite matchable data set includes: data generation time: "wind_result_COG_time"; bottom height of the Bin (bottom height of the data profile): "wind_result_bottom_altitude"; top height of the Bin (top height of the data profile): "wind_result_top_altitude"; latitude of the data profile: "wind_result_COG_latitude"; longitude of the data profile: "wind_result_COG_longitude"; product horizontal line-of-sight wind speed error: "wind_result_HLOS_error"; wind speed: "wind_result_wind_velocity"; wind speed type: "wind_result_observation_type"; wind speed validity flag: "wind_result_validity_flag"; wind speed line-of-sight angle: "wind_result_los_azimuth", etc.

[0110] Step 11: If the Aeolus data parsing fails, end the current judgment and return to Step 3 to continue obtaining the satellite matchable data set for the next site.

[0111] In summary, the process of obtaining the satellite matchable data set is a process of determining the satellite wind profile data that can be matched based on the current ground-based site, that is, the Aeolus satellite data preparation process, for the inversion of the spaceborne lidar wind field UV wind vector and the construction of the wind speed and direction inversion algorithm model.

[0112] In summary, the present disclosure overcomes the ill-posed problem of estimating multiple unknown variables (horizontal wind zonal wind, meridional wind components) from a single remote sensing observable, i.e., horizontal line-of-sight wind. By taking advantage of the machine learning algorithm, an inversion model is established, and then through the construction of a paired data set for training, the goal of obtaining a set of horizontal wind zonal wind and meridional wind components by inputting a single horizontal line-of-sight wind is finally achieved.

[0113] According to the embodiments of the present disclosure, the following technical effects are achieved:

[0114] It is possible to obtain the optimal paired data of the spaceborne lidar wind field, including satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, the site longitude and latitude, the satellite azimuth angle, and the wind speed and direction of the radiosonde. Then, input the optimal paired data into the constructed wind speed and direction inversion algorithm model to output the UV wind vector. Based on this, a spaceborne lidar Doppler radar wind field inversion technology based on machine learning uses various machine learning methods to establish a spaceborne lidar Doppler radar wind field inversion model, that is, a wind speed and direction inversion algorithm model, and successfully realizes the function of inversing the UV wind vector from the horizontal line-of-sight wind observation, that is, the zonal wind and meridional wind components.

[0115] In some embodiments, the above method further includes:

[0116] Convert the UV wind vector into wind speed and direction;

[0117] Based on preset evaluation indicators and preset evaluation dimensions, evaluate the wind speed and direction respectively against the wind profile data, ERA5 data, and Beidou radiosonde data. The preset evaluation indicators include correlation coefficient, root mean square error, mean absolute error, and mean deviation, and the preset evaluation dimensions include site dimension, height layer dimension, month dimension, and overall dimension.

[0118] In some embodiments, the inversed UV wind can be converted into wind speed and direction and evaluated against the corresponding data such as ground-based wind profile radar, ERA5, and Beidou radiosonde. In the evaluation stage, evaluation dimensions and evaluation indicators can be set. For example, correlation coefficient, root mean square error, mean deviation, mean absolute error, etc. are used as evaluation indicators to generate evaluation results for different sites, different height layers, and different months, so as to further evaluate the inversion results of the spaceborne lidar wind field UV wind vector and improve the inversion accuracy.

[0119] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present disclosure is not limited by the described action sequence, because according to the present disclosure, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present disclosure.

[0120] The above is the introduction of the method embodiments. The following further illustrates the solution of the present disclosure through device embodiments.

[0121] Figure 2 The block diagram of an inversion device 200 for the spaceborne lidar wind field UV wind vector according to an embodiment of the present disclosure is shown. AsFigure 2 As shown in the figure, the device 200 includes:

[0122] An acquisition module 210, configured to acquire the optimal paired data of the spaceborne lidar wind field; the optimal paired data includes the satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, the site longitude and latitude, the satellite azimuth angle, and the wind speed and wind direction of the radiosonde; the optimal paired data is determined according to the satellite data, ERA5 data, Beidou radiosonde data, and ground-based wind profile radar data;

[0123] An inversion module 220, configured to input the optimal paired data into the constructed wind speed and wind direction inversion algorithm model, and output the UV wind vector. The wind speed and wind direction inversion algorithm model uses the support vector machine regression algorithm and the min-max normalization model for data inversion.

[0124] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the described module can refer to the corresponding process in the foregoing method embodiment, and will not be elaborated herein.

[0125] In the technical solution of the present disclosure, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0126] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0127] Figure 3 The block diagram of an exemplary electronic device 300 capable of implementing the embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0128] The electronic device 300 includes a computing unit 301, which can execute various appropriate actions and processes according to the computer program stored in the ROM 302 or the computer program loaded from the storage unit 308 into the RAM 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 can also be stored. The computing unit 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. The I / O interface 305 is also connected to the bus 304.

[0129] Multiple components in the electronic device 300 are connected to the I / O interface 305, including: an input unit 306, such as a keyboard, a mouse, etc.; an output unit 307, such as various types of displays, speakers, etc.; a storage unit 308, such as a disk, an optical disc, etc.; and a communication unit 309, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 309 allows the electronic device 300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0130] The computing unit 301 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 301 executes the various methods and processes described above, such as method 100. For example, in some embodiments, method 100 can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 308.

[0131] In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 300 via the ROM 302 and / or the communication unit 309. When the computer program is loaded into the RAM 303 and executed by the computing unit 301, one or more steps of the method 100 described above can be executed. Alternatively, in other embodiments, the computing unit 301 can be configured to execute method 100 in any other suitable manner (e.g., by means of firmware).

[0132] The various embodiments of the systems and technologies described above in this article can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs, which can be executed and / or interpreted on a programmable system including at least one programmable processor, and the programmable processor can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0133] The program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, a special purpose computer, or other programmable data processing device, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program codes may be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0134] In the context of the present disclosure, a machine-readable medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0135] In order to provide interaction with a user, the systems and techniques described herein may be implemented on a computer having: a display device for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0136] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0137] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, a server of a distributed system, or a server incorporating blockchain.

[0138] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this is not limited herein.

[0139] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A method for retrieving the UV wind vector of the wind field by spaceborne lidar, characterized in that, Including: Obtaining the optimal paired data of the spaceborne lidar wind field; the optimal paired data includes the satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, the site longitude and latitude, the satellite azimuth angle, and the wind speed and wind direction of the radiosonde; the optimal paired data is determined according to satellite data, ERA5 data, Beidou radiosonde data, and ground-based wind profile radar data; Inputting the optimal paired data into the constructed wind speed and wind direction inversion algorithm model, and outputting the UV wind vector. The wind speed and wind direction inversion algorithm model uses the support vector machine regression algorithm and the min-max normalization model for data inversion.

2. The method according to claim 1, characterized in that The construction of the wind speed and wind direction inversion algorithm model includes: Obtaining historical optimal paired data, which includes historical satellite horizontal line-of-sight wind speed, historical top height of the data profile, historical bottom height of the data profile, historical site longitude and latitude, historical satellite azimuth angle, historical wind speed and wind direction of the radiosonde, and the UV wind vector of historical ERA5; Based on the preset min-max normalization model, preprocessing the historical optimal paired data; Based on the support vector machine regression algorithm, using the preprocessed historical satellite horizontal line-of-sight wind speed, historical top height of the data profile, historical bottom height of the data profile, historical site longitude and latitude, historical satellite azimuth angle, and historical wind speed and wind direction of the radiosonde as samples, and the UV wind vector of historical ERA5 as labels, training the preset Fengshen wind speed and wind direction inversion algorithm model with the obtained training set to obtain the constructed wind speed and wind direction inversion algorithm model; Among them, the training of the preset Fengshen wind speed and wind direction inversion algorithm model includes: finding the optimal parameters of the preset Fengshen wind speed and wind direction inversion algorithm model based on grid search and cross-validation, and using the optimal parameters to train the preset Fengshen wind speed and wind direction inversion algorithm model.

3. The method according to claim 2, wherein The obtaining of the historical optimal paired data includes: Obtaining the satellite matchable data set and the ERA5 and ground-based data match data set; According to the preset corresponding relationship between the satellite-ERA5 and ground-based data height layers, converting the wind speed and wind direction in the multi-layer ERA5 and ground-based data match data corresponding to the height layer of the satellite matchable data set into UV wind vectors; Arithmetically averaging the UV wind vectors to obtain the mean value of the UV wind vectors; Converting the mean value of the UV wind vectors into the average wind speed and wind direction; Based on the preset satellite wind speed conversion algorithm, calculating the satellite horizontal line-of-sight wind speed according to the average wind speed and wind direction; According to the satellite horizontal line-of-sight wind speed, matching the satellite matchable data set and the ERA5 and ground-based data match data set to obtain the historical optimal paired data.

4. The method according to claim 3, wherein Obtaining the ERA5 and ground-based data match data set includes: Obtaining ERA5 data and ground-based data, where the ground-based data includes Beidou radiosonde data and ground-based wind profile radar data; According to the preset screening rules for the matchable data between Beidou radiosonde and ERA5, obtain the matchable dataset of ERA5 and Beidou radiosonde data; wherein, the preset screening rules for the matchable data between Beidou radiosonde and ERA5 include: the ERA5 data meets the requirement of the preset spatial distance from the Beidou radiosonde ground station, and the Beidou radiosonde data and the ERA5 data are within the first preset nearest time and the preset pressure difference. According to the preset screening rules for the matchable data between ground-based wind profile and ERA5, obtain the matchable dataset of ERA5 and ground-based data; wherein, the preset screening rules for the matchable data between ground-based wind profile and ERA5 include: the matchable dataset of ERA5 and Beidou radiosonde data and the ERA5 data are within the second preset nearest time and the preset height difference.

5. The method according to claim 4, wherein The obtaining of the ERA5 data and the ground-based data includes: When the ERA5 data meets the requirement of the preset spatial distance from the ground-based station, obtain the corresponding ERA5 data. Based on the ERA5 data, obtain the Beidou radiosonde data and the ground-based wind profile radar data within the preset time.

6. The method according to claim 5, characterized in that, The obtaining of the satellite matchable dataset includes: According to the preset screening rules for satellite matchable data, obtain the satellite matchable dataset; wherein, the preset screening rules for satellite matchable data include: the time of the satellite wind profile data is within the preset matching time period, the longitude and latitude of the data profile of the satellite wind profile data are within the preset longitude and latitude range of the ground-based station, and the satellite passes over the ground-based station.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Convert the UV wind vector into wind speed and wind direction; Based on the preset evaluation indicators and preset evaluation dimensions, evaluate the wind speed and wind direction respectively with the wind profile data, ERA5 data, and Beidou radiosonde data; the preset evaluation indicators include correlation coefficient, root mean square error, mean absolute error, and mean deviation, and the preset evaluation dimensions include station dimension, height layer dimension, month dimension, and integrity dimension.

8. An inversion device for the UV wind vector of the wind field of a spaceborne lidar, characterized in that, It includes: An acquisition module, configured to acquire the optimal paired data of the spaceborne lidar wind field; the optimal paired data includes satellite horizontal line-of-sight wind speed, the top height of the data profile, the bottom height of the data profile, station longitude and latitude, satellite azimuth, and the wind speed and wind direction of the radiosonde; the optimal paired data is determined according to satellite data, ERA5 data, Beidou radiosonde data, and ground-based wind profile radar data; An inversion module, configured to input the optimal paired data into the constructed wind speed and wind direction inversion algorithm model, and output the UV wind vector, and the wind speed and wind direction inversion algorithm model uses the support vector machine regression algorithm and the min-max normalization model for data inversion.

9. An electronic device, characterized in that, It includes: At least one processor; And A memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-7.

Images