Global lake wind speed estimation method based on satellite-borne BDS / GPS / GLONASS / Galileo reflected signals of China Tangyu No.1

Abstract

The invention discloses a global lake wind speed estimation method based on a BDS / GPS / GLONASS / Galileo reflected signal of a China Tangyu No.1 satellite-borne, which comprises three modules, the global lake data extraction module is based on reflection measurement of four systems (BDS / GPS / GLONASS / Galileo) of a temmoku No.1 constellation, the module is used for screening relevant important feature variables of global lake wind speed inversion through a machine learning algorithm and calculating input feature importance values through an SHAP, and the module is used for constructing a global lake water surface wind speed inversion model. According to the method, the quantitative result of global lake wind speed inversion through satellite-borne BDS / GPS / GLONASS / Galileo reflected signals of the Tangyu No.1 constellation is achieved for the first time, the lake wind speed inversion precision is improved by adopting a machine learning method, and the huge potential of satellite-borne GNSS-R in inland water body wind speed monitoring is powerfully promoted.

Description

Technical Field

[0001] The present invention belongs to the technical field of cross-research on lake wind speed parameter inversion of GNSS reflection signal and machine learning, and specifically, relates to a method for estimating global lake wind speed using BDS / GPS / GLONASS / Galileo reflection signals onboard China's Tianmu-1 satellite. Background Art

[0002] As an important freshwater resource, accurate measurement of surface wind speed in lakes is of great significance for key areas such as meteorological forecasting, environmental monitoring and water resource management. It helps to improve the accuracy of climate and weather forecasts, monitor the dynamics of lake ecosystems, and assess wind energy potential. At present, methods for measuring surface wind speed in inland water bodies such as lakes include obtaining real-time data from meteorological stations, anemometers on buoys and wind vanes. Radar (especially synthetic aperture radar, SAR) can collect data through ground or space platforms (such as satellites and aircraft). Therefore, radar technology can be used for both satellite remote sensing and ground or aerial remote sensing. It has the advantages of real-time monitoring, wide coverage and strong adaptability in monitoring surface wind speed in lakes, but it also has disadvantages such as high cost, limited accuracy and electromagnetic interference. In addition, unmanned aerial vehicles (UAVs) carrying sensors flying over lakes to collect wind speed data is also a flexible monitoring method. GNSS-R technology captures the signals emitted by the global navigation satellite system and reflected by the lake surface, processes and analyzes the delay and Doppler frequency of these signals, and generates delay Doppler maps (DDM) to extract features related to wind speed. By using the correlation between wind speed and water surface roughness, combined with physical or empirical models, GNSS-R can convert these characteristics into an estimate of lake surface wind speed. Although GNSS-R technology is relatively mature in the inversion of sea surface wind speed, most current studies are limited to using single-system, dual-system or triple-system reflection signal observation data and only using a limited number of space-borne GNSS-R satellites to invert sea surface wind speed. At present, there are no quantitative experimental results for lake wind speed inversion using GNSS-R, especially space-borne multiple GNSS-R satellites.

[0003] In view of this, the present invention is proposed. Summary of the invention

[0004] In order to achieve the first quantitative results of global lake wind speed inversion of four satellite-borne GNSS-R systems (BDS / GPS / GLONASS / Galileo), the present invention proposes for the first time a method for estimating global lake wind speed by using BDS / GPS / GLONASS / Galileo reflection signals carried by the Chinese Tianmu-1 constellation composed of 22 low-orbit small satellites. The present invention provides a method for estimating global lake wind speed by using BDS / GPS / GLONASS / Galileo reflection signals carried by the Chinese Tianmu-1 constellation. The method mainly includes three modules, namely a global lake data extraction module based on reflection measurement of the four systems (BDS / GPS / GLONASS / Galileo) of the Tianmu-1 constellation, a module for screening important characteristic variables related to global lake wind speed inversion by machine learning algorithm and calculating input characteristic importance values ​​by SHAP, and a module for building a global lake surface wind speed inversion model. Among them, the machine learning algorithm screens the important characteristic variables related to the global lake wind speed inversion and the SHAP calculation input characteristic importance value module, uses the bagged tree model to screen the characteristic variables used to build the lake surface wind speed model, and uses SHAP to calculate the input characteristic importance value, so as to better understand the relationship between the characteristic variables and the lake wind speed inversion model, and then improve the accuracy of lake wind speed inversion. The present invention has for the first time achieved the quantitative results of the global lake wind speed inversion inversion of the BDS / GPS / GLONASS / Galileo reflection signal on the Tianmu-1 constellation, and used machine learning methods to improve the accuracy of lake wind speed inversion, which has strongly promoted the huge potential of satellite-borne GNSS-R in inland water wind speed monitoring.

[0005] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0006] A method for estimating global lake wind speed from BDS / GPS / GLONASS / Galileo reflected signals onboard China's Tianmu-1 satellite comprises the following steps:

[0007] Step S1, obtaining the reflected signal data of the four systems (BDS, GPS, GLONASS and Galileo) of the Tianmu-1 constellation (provided free of charge by Aerospace Tianmu (Chongqing) Satellite Technology Co., Ltd. and Aerospace Science and Industry (Beijing) Space Information Application Co., Ltd.), ERA5 lake surface wind speed, temperature, water depth, lake ice thickness and rainfall data, and CCMP lake wind speed data;

[0008] Step S2, extracting the Tianmu-1 reflection data observations, selecting ERA5 lake surface temperature, water depth, lake ice thickness and rainfall auxiliary data variables, and performing spatiotemporal matching on all data sets;

[0009] Step S3, extracting global lake data using the global lake shp file, and filtering out lake data with ice thickness ≤ 0.05 m according to the lake ice thickness auxiliary parameters in the data set;

[0010] Step S4, data preprocessing and data quality control filtering, and dividing the training data set, validation data set and test data set, accounting for 60%, 30% and 10% of the total data set respectively;

[0011] Step S5, using a machine learning algorithm to screen the characteristic variables used to construct the lake surface wind speed model, and using the SHAP (SHapley Additive exPlanations) interpretability framework to calculate the characteristic importance value (SHAP value) of each model input parameter.

[0012] Step S6, construct a global lake wind speed machine learning inversion model based on BDS / GPS / GLONASS / Galileo reflection signals, and use the training data set and test data set to train the lake wind speed inversion models of different systems and test the inversion wind speed performance of different systems.

[0013] Furthermore, the onboard GNSS-R data uses the reflected signal data of the four systems (BDS, GPS, GLONASS and Galileo) of the Tianmu-1 constellation provided free of charge by Aerospace Tianmu (Chongqing) Satellite Technology Co., Ltd. and Aerospace Science and Industry (Beijing) Space Information Application Co., Ltd. The reference wind speed product uses the ERA5 lake surface wind speed product and CCMP lake wind speed product. The external auxiliary data uses the ERA5 lake temperature, water depth, lake ice thickness and rainfall data.

[0014] Furthermore, the Tianmu-1 reflection data observations and auxiliary data variables are merged. In order to avoid the influence of wind speed inversion errors caused by spatiotemporal differences in different data sets, all auxiliary data sets (ERA5 and CCMP) are spatiotemporally matched with the Tianmu-1 data, and the spatiotemporal resolution is unified to obtain the matched data sets. In terms of time, ERA5 and CCMP are aligned to the Coordinated Universal Time (UTC). In terms of space, two-dimensional linear interpolation is used to perform spatial interpolation of Tianmu-1 data and two auxiliary data sets (ERA5 and CCMP). In addition, for the reference wind speeds used, the wind speeds extracted from CCMP and ERA5 are both u wind components and v wind components, so the optimal wind speed calculation must be performed first to obtain the true wind speed. The calculation formula is:

[0015]

[0016] Where WS is the true wind speed, u and v are the eastward and northward components of the wind speed, respectively.

[0017] Furthermore, the global lake shp file is used to extract global lake data, and the lake ice thickness will change the scattering characteristics of the signal on the water surface, causing signal attenuation. The extracted lake data are filtered out according to the lake ice thickness auxiliary parameters in the data set to select lake ice thickness ≤ 0.05m.

[0018] Furthermore, for the data set, the variable value of -9999.9 is assigned to NaN and removed, and then the data set is normalized. After data filtering, the data set is divided into training set, test set and validation set in a ratio of 6:3:1.

[0019] Further, in step S5, the XGBoost-based Bagging regression model is used to screen important feature variables for constructing the lake surface wind speed model (the feature variables used for screening are shown in Table 1), wherein the learning rate of the model is set to 0.02; the basic learner type selects a tree-based model; the maximum number of leaf nodes is set to 127; in order to prevent overfitting, the feature ratio of each tree randomly sampled is set to 0.6. The SHAP (SHapley Additive exPlanations) interpretability framework is used to calculate the feature importance values ​​(SHAP values) of the top ten important feature variables (Sp_lon, Sp_alt, Sp_lat, Rx_lat, temperature, Ddm_sp_reflectivity, Sample_num, precipitation, Direct_signal_noise, Sp_fresnel_coeff_square) selected. To interpret the model in SHAP, you need to create an explainer first, and the SHAP explainer used is Kernel. The specific calculation formula of the SHAP value is as follows:

[0020]

[0021] In the formula, φ i is the contribution of feature i to the model output (i.e., the SHAP value), N is the set of all features, S is the feature subset excluding feature i, f(S) is the model prediction value on feature set S, f(S∪{i}) is the model prediction value on feature S and feature i, |S|! and (|N|-|S|-1)! are the factorial weights of the feature permutations.

[0022] Table 1. Feature variables screened by machine learning

[0023]

[0024]

[0025]

[0026] Furthermore, step S6: construct a bagging tree (BT) machine learning model for global lake surface wind speed inversion, input the training set into the model training, use the validation set to supervise the model training to prevent overfitting, and use the test set to test the model to obtain the inverted wind speed. Four accuracy evaluation indicators (root mean square error (RMSE), bias, mean absolute error (MAE), and correlation coefficient (CC)) are used to comprehensively evaluate the predicted wind speed of the lake surface wind speed inversion model constructed based on the reflection signal observation data of the four systems.

[0027] After adopting the above technical scheme, the present invention has the following beneficial effects compared with the prior art.

[0028] The present invention uses the reflected signal data of the four satellite-borne GNSS-R systems (BDS / GPS / GLONASS / Galileo) for the first time to invert the global lake wind speed, and obtains reliable accuracy. In addition, the present invention uses a machine learning model to screen the characteristic variables used to construct the lake surface wind speed model, which effectively improves the accuracy of lake wind speed inversion compared to manually extracting characteristic variables.

[0029] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are part of this application and are used to provide a further understanding of the present invention. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the drawings described below are only some embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0031] Figure 1 Schematic diagram of the method flow in an embodiment of the present invention.

[0032] Figure 2 This is a diagram showing the importance order of characteristic variables related to lake wind speed inversion in an embodiment of the present invention.

[0033] Figure 3 This is a scatter density map of global lake wind speeds and ERA5 reference wind speeds for different systems inverted by the manually screened feature input model in an embodiment of the present invention.

[0034] Figure 4 This is a scatter density map of global lake wind speeds and CCMP reference wind speeds for different systems inverted by the manually screened feature input model in an embodiment of the present invention.

[0035] Figure 5This is a scatter density map of global lake wind speeds and ERA5 reference wind speeds for different systems inverted by the machine learning screening feature input model in an embodiment of the present invention.

[0036] Figure 6 This is a scatter density map of global lake wind speeds and CCMP reference wind speeds for different systems inverted by the machine learning screening feature input model in an embodiment of the present invention.

[0037] Figure 7 This is a comparison diagram of the importance of global lake wind speed inversion features in the implementation cases of the present invention in the embodiments of the present invention.

[0038] Figure 8 This is a global deviation distribution map of global lake wind speeds in different systems inverted by the machine learning screening feature input model in an embodiment of the present invention and the ERA5 reference wind speed.

[0039] Fig. 9 This is a global deviation distribution map of global lake wind speeds in different systems and CCMP reference wind speeds inverted by the machine learning screening feature input model in an embodiment of the present invention.

[0040] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0042] Example 1

[0043] In order to verify the effectiveness of the proposed method, the GNSS-R observation data of the four systems of the Tianmu-1 constellation from 2024.03.11 to 2024.03.31, the ERA5 lake surface wind speed, temperature, water depth, lake ice thickness and rainfall data, and the CCMP lake wind speed data were obtained for experiments.

[0044] Combined with Figure 1-9 The method for estimating global lake wind speed using BDS / GPS / GLONASS / Galileo reflected signals carried by China's Tianmu-1 satellite in this embodiment includes:

[0045] Step S1, obtaining the reflected signal data of the four systems (BDS, GPS, GLONASS and Galileo) of the Tianmu-1 constellation (provided free of charge by Aerospace Tianmu (Chongqing) Satellite Technology Co., Ltd. and Aerospace Science and Industry (Beijing) Space Information Application Co., Ltd.), ERA5 lake surface wind speed, temperature, water depth, lake ice thickness and rainfall data, and CCMP lake wind speed data;

[0046] Step S2, extracting the Tianmu-1 reflection data observations, selecting ERA5 lake surface temperature, water depth, lake ice thickness and rainfall auxiliary data variables, and performing spatiotemporal matching on all data sets;

[0047] Step S3, extracting global lake data using the global lake shp file, and filtering out lake data with ice thickness ≤ 0.05 m according to the lake ice thickness auxiliary parameters in the data set;

[0048] Step S4, data preprocessing and data quality control filtering, and dividing the training data set, validation data set and test data set, accounting for 60%, 30% and 10% of the total data set respectively;

[0049] Step S5, using a machine learning algorithm to screen the characteristic variables used to construct the lake surface wind speed model, and using the SHAP (SHapley Additive exPlanations) interpretability framework to calculate the characteristic importance value (SHAP value) of each model input parameter.

[0050] Step S6, construct a global lake wind speed machine learning inversion model based on BDS / GPS / GLONASS / Galileo reflection signals, and use the training data set and test data set to train the lake wind speed inversion models of different systems and test the inversion wind speed performance of different systems.

[0051] As an implementation of this embodiment, in step S1, the satellite-borne GNSS-R data uses the reflection signal data of the four systems (BDS, GPS, GLONASS and Galileo) of the Tianmu-1 constellation provided free of charge by Aerospace Tianmu (Chongqing) Satellite Technology Co., Ltd. and Aerospace Science and Industry (Beijing) Space Information Application Co., Ltd., and the reference wind speed product uses the ERA5 lake surface wind speed product and CCMP lake wind speed product. The external auxiliary data uses the ERA5 lake temperature, water depth, lake ice thickness and rainfall data.

[0052] As an implementation method of this embodiment, in step S2, first, the Tianmu-1 reflection data observations and auxiliary data variables are merged. In order to avoid the influence of the spatiotemporal differences on different data sets on the wind speed inversion error, all auxiliary data sets (ERA5 and CCMP) are spatiotemporally matched with the Tianmu-1 data, and the spatiotemporal resolution is unified to obtain the matched data set. The specific implementation method is: in terms of time, ERA5 and CCMP are aligned to the Coordinated Universal Time (UTC); in space, two-dimensional linear interpolation is used to perform spatial interpolation of Tianmu-1 data and two auxiliary data sets (ERA5 and CCMP). In addition, for the reference wind speed used in the embodiment, the wind speeds extracted from CCMP and ERA5 are both u wind components and v wind components, so the optimal wind speed calculation must be performed first to obtain the true wind speed, and the calculation formula is:

[0053]

[0054] Where WS is the true wind speed, u and v are the eastward and northward components of the wind speed, respectively.

[0055] As an implementation of this embodiment, in step S3, the global lake shp file is used to extract global lake data, and the lake ice thickness will change the scattering characteristics of the signal on the water surface, causing signal attenuation. The extracted lake data is filtered out according to the lake ice thickness auxiliary parameter in the data set to obtain lake ice thickness ≤ 0.05m.

[0056] As an implementation of this embodiment, in step S4, for the data set, the variable value -9999.9 is assigned to NaN and removed, and then the data set is normalized. After data filtering, the data set is divided into a training set, a test set, and a validation set in a ratio of 6:3:1.

[0057] As an implementation method of this embodiment, in step S5, the XGBoost-based Bagging regression model is used to screen important feature variables for constructing a lake surface wind speed model (the feature variables used for screening are shown in Table 1), wherein the learning rate of the model is set to 0.02; the basic learner type selects a tree-based model; the maximum number of leaf nodes is set to 127; in order to prevent overfitting, the feature ratio of each tree randomly sampled is set to 0.6. The SHAP (SHapleyAdditive exPlanations) interpretability framework is used to calculate the feature importance values ​​(SHAP values) of the top ten important feature variables (Sp_lon, Sp_alt, Sp_lat, Rx_lat, temperature, Ddm_sp_reflectivity, Sample_num, precipitation, Direct_signal_noise, Sp_fresnel_coeff_square) screened out. To interpret the model in SHAP, you need to first create an explainer (explainer), and the SHAP explainer used is Kernel. The specific calculation formula of the SHAP value is as follows:

[0058]

[0059] In the formula, φ i is the contribution of feature i to the model output (i.e., the SHAP value), N is the set of all features, S is the feature subset excluding feature i, f(S) is the model prediction value on feature set S, f(S∪{i}) is the model prediction value on feature S and feature i, |S|! and (|N|-|S|-1)! are the factorial weights of the feature permutations.

[0060] Table 1. Feature variables screened by machine learning

[0061]

[0062]

[0063]

[0064] Attached Figure 7 This is a comparison chart of the importance of wind speed inversion features in lakes around the world. The analysis shows that:

[0065] From the attached Figure 7The upper left figure shows that features such as Sp_lon, Sp_alt, and Sp_lat have large positive SHAP values, which greatly contribute to the positive prediction of the model; temperature and Ddm_sp_reflectivity have both positive and negative contributions, indicating that high and low values ​​of these features will affect the model in different directions; Direct_signal_noise and Sp_fresnel_coeff_square are more concentrated in the negative contribution area, which means that high values ​​of these features may reduce the prediction results of the model. Figure 7 The middle figure shows that Sp_lon is the most important feature, which has the greatest impact on model prediction, followed by Sp_alt and Sp_lat, which also contribute relatively more to the model; temperature, Ddm_sp_reflectivity and Sample_num are also relatively important features in model prediction. Figure 7 The upper right corner of the figure shows that Sp_lon has the largest positive contribution to the sample, pushing up the predicted value by 0.29; Sp_alt also has a large positive contribution, increasing by 0.26; Rx_lat has a negative contribution, decreasing by 0.11. Figure 7 As can be seen in the figure below, Sp_alt and Sp_lon are the two most important positive contribution features, pushing up the predicted value by 0.26 and 0.29 respectively. Rx_lat has a negative impact on the predicted value, reducing it by 0.11; other features such as Sp_lat, Sp_fresnel_coeff_square, etc. also make corresponding contributions to the prediction results.

[0066] As an implementation method of this embodiment, step S6 describes constructing a global lake wind speed machine learning inversion model based on BDS / GPS / GLONASS / Galileo reflection signals, using a bagging tree (Bagging Tree) machine learning model to construct a global lake surface wind speed inversion model, inputting a training set into model training, using a validation set to supervise model training to prevent overfitting, and using a test set to test the model to obtain the inverted wind speed. Four accuracy evaluation indicators (root mean square error (RMSE), bias, mean absolute error (MAE), and correlation coefficient (CC)) are used to comprehensively evaluate the predicted wind speed of the lake surface wind speed inversion model constructed based on the four system reflection signal observation data.

[0067] The following is a comprehensive evaluation of the four systems' predicted lake wind speeds: Figure 3The scatter density plots of global lake wind speeds of different systems inverted by the model with manually selected features (60, as shown in Table 1) and the ERA5 reference wind speed are shown. The accuracy evaluation results of the four systems (BDS, GPS, GAL, GLO) are RMSE: 1.506m / s, 1.877m / s, 1.662m / s, 1.612m / s, CC: 0.74, 0.65, 0.64, 0.7, MAE: 0.19m / s, 0.556m / s, 0.19m / s, 0.357m / s. Figure 5 The scatter density diagram of global lake wind speeds of different systems inverted by the machine learning screening features (10) input model and the ERA5 reference wind speed is shown. The accuracy evaluation results of the four systems (BDS, GPS, GAL, GLO) are RMSE: 1.237m / s, 1.569m / s, 1.488m / s, 1.527m / s, CC: 0.84, 0.77, 0.72, 0.73, MAE: 0.099m / s, 0.482m / s, 0.073m / s, 0.213m / s. By comparing the attached Figure 3 and attached Figure 5 It can be seen that by using machine learning to screen the features required for the inversion of global lake wind speed, the inverted wind speeds of the four systems have improved in RMSE, MAE and CC. Figure 4 and attached Figure 6 The corresponding scatter density plots are the global lake wind speeds of different systems and the CCMP reference wind speeds inverted by the input models of manual screening features and machine learning screening features. Figure 8 and attached Fig. 9 The global deviation distribution maps of global lake wind speeds in different systems inverted by machine learning feature input models and ERA5 and CCMP reference wind speeds are shown in Figure 2. Figure 8 and attached Fig. 9 It can be seen that the deviations of the wind speed inversion results of the four systems of Tianmu-1 from the ERA5 reference wind speed are generally better than the deviations from the CCMP reference wind speed. In summary, the accuracy of wind speed inversion has been significantly improved by using machine learning methods to screen characteristic variables. Further analysis shows that the comparison accuracy of the wind speed inversion results of the four systems of Tianmu-1 with the ERA5 reference wind speed is generally better than the comparison accuracy with the CCMP reference wind speed. In addition, among the four systems of the Tianmu-1 constellation, the wind speed inversion results of the BDS system are generally better than those of the GPS, GAL and GLO systems.

[0068] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment, it is not used to limit the present invention. Any technician familiar with this patent can make some changes or modify the technical contents suggested above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the solution of the present invention.

Claims

1. A method for estimating global lake wind speed from BDS / GPS / GLONASS / Galileo reflected signals onboard China's Tianmu-1 satellite, characterized in that: The following steps are involved: Step S1, obtaining the reflected signal data of the four systems (BDS, GPS, GLONASS and Galileo) of the Tianmu-1 constellation, the ERA5 lake surface wind speed, temperature, water depth, lake ice thickness and rainfall data, and the CCMP lake wind speed data; Step S2, extracting the Tianmu-1 reflection data observations, selecting ERA5 lake surface temperature, water depth, lake ice thickness and rainfall auxiliary variable data, and performing spatiotemporal matching on all data sets; Step S3, using the global lake shp file to extract global lake GNSS-R observation data, and filtering out lake ice thickness ≤ 0.05 m data according to lake ice thickness auxiliary parameters in the data set; Step S4, data preprocessing and data quality control filtering, and dividing the training data set, validation data set and test data set, accounting for 60%, 30% and 10% of the total data set respectively; Step S5, using a machine learning algorithm to screen characteristic variables for constructing a lake surface wind speed model, and using the SHAP interpretability framework to calculate the characteristic importance value (SHAP value) of each model input parameter; Step S6, construct a global lake wind speed machine learning inversion model based on BDS / GPS / GLONASS / Galileo reflection signals, and use the training data set and test data set to train the lake wind speed inversion models of different systems and test the inversion wind speed performance of different systems.

2. A method for estimating global lake wind speed based on BDS / GPS / GLONASS / Galileo reflected signals from China's Tianmu-1 satellite according to claim 1, characterized in that: In step S1, the onboard GNSS-R data adopts the reflected signal data of the four systems (BDS, GPS, GLONASS and Galileo) of the Tianmu-1 constellation provided free of charge by Aerospace Tianmu (Chongqing) Satellite Technology Co., Ltd. and Aerospace Science and Industry (Beijing) Space Information Application Co., Ltd., the reference wind speed product adopts the ERA5 lake surface wind speed product and CCMP lake wind speed product, and the external auxiliary data adopts the ERA5 lake temperature, water depth, lake ice thickness and rainfall data.

3. A method for estimating global lake wind speed based on BDS / GPS / GLONASS / Galileo reflected signals from China's Tianmu-1 satellite according to claim 1, characterized in that: In step S2, the Tianmu-1 reflection data observations and auxiliary data variables are first merged, and then all auxiliary data sets are temporally and spatially matched with the Tianmu-1 data to unify the temporal and spatial resolutions to obtain the matched data sets. The specific implementation is: In terms of time, ERA5 and CCMP are aligned to Coordinated Universal Time (UTC); in terms of space, two-dimensional linear interpolation is used to spatially interpolate the Tianmu-1 data and two auxiliary data sets (ERA5 and CCMP); in addition, for the reference wind speed used, the wind speeds extracted from CCMP and ERA5 are u wind components and v wind components, so the optimal wind speed calculation must be performed first to obtain the true wind speed, and the calculation formula is: Where WS is the true wind speed, u and v are the eastward and northward components of the wind speed, respectively.

4. A method for estimating global lake wind speed based on BDS / GPS / GLONASS / Galileo reflected signals from China's Tianmu-1 satellite according to claim 1, characterized in that: Step S4 is specifically as follows: for the data set, the variable value of -9999.9 is assigned to NaN and removed, and then the data set is normalized. After data filtering, the data set is divided into a training set, a test set, and a validation set in a ratio of 6:3:

1.

5. A method for estimating global lake wind speed based on BDS / GPS / GLONASS / Galileo reflected signals from China's Tianmu-1 satellite according to claim 1, characterized in that: Step S5 is specifically as follows: using the Bagging regression model of XGBoost to screen important feature variables for constructing the lake surface wind speed model from 60 variables, wherein the learning rate of the model is set to 0.02; the basic learner type selects a tree-based model; the maximum number of leaf nodes is set to 127; in order to prevent overfitting, the feature ratio of each tree randomly sampled is set to 0.6; using the SHAP (SHapley Additive exPlanations) interpretability framework, the feature importance values ​​(SHAP values) of the top ten important feature variables (Sp_lon, Sp_alt, Sp_lat, Rx_lat, temperature, Ddm_sp_reflectivity, Sample_num, precipitation, Direct_signal_noise, Sp_fresnel_coeff_square) selected are calculated. To interpret the model in SHAP, an explainer needs to be created first. The explainer of SHAP used is Kernel, and the specific calculation formula of the SHAP value is as follows: In the formula, φ i is the contribution of feature i to the model output (i.e., the SHAP value), N is the set of all features, S is the feature subset excluding feature i, f(S) is the model prediction value on feature set S, f(S∪{i}) is the model prediction value on feature S and feature i, |S|! and (|N|-|S|-1)! are the factorial weights of the feature permutations.

6. A method for estimating global lake wind speed based on BDS / GPS / GLONASS / Galileo reflected signals onboard China Tianmu-1 according to claim 1, characterized in that: Step S6 is specifically as follows: construct a bagging tree (BT) machine learning model for global lake surface wind speed inversion, input the training set into the model training, use the validation set to supervise the model training to prevent overfitting, and use the test set to test the model to obtain the inverted wind speed. Four accuracy evaluation indicators (root mean square error (RMSE), bias, mean absolute error (MAE), and correlation coefficient (CC)) are used to comprehensively evaluate the predicted wind speed of the lake surface wind speed inversion model constructed based on the reflection signal observation data of the four systems.

Images