Multi-source satellite offshore area wind resource atlas analysis method, device and equipment

Abstract

The invention provides a multi-source satellite offshore area wind resource atlas analysis method, device and equipment, and the method comprises the steps: carrying out the preprocessing of a multi-source original sea surface wind field, so as to determine a multi-source target sea surface wind field corresponding to a first grid; constructing an average sea surface wind field set based on the wind measurement station data, the multi-source target sea surface wind field corresponding to the first grid and the reanalysis data; training a machine learning model by using the average sea surface wind field set and the sea surface feature data so as to predict a satellite average sea surface wind field of the second grid based on the reanalysis data corresponding to the second grid through the trained machine learning model; and generating a wind resource map corresponding to the research sea area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid. According to the method, the integrity of regional data, the consistency with actual observation and the data accuracy can be effectively ensured, and the difficulty of collecting data in wind resource evaluation and prediction can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of microwave remote sensing, and in particular to a method, device and equipment for analyzing multi-source satellite offshore regional wind resource maps. Background Art

[0002] With the acceleration of the industrialization process, the demand for renewable energy is increasing continuously. As a renewable and pollution-free energy, wind energy is widely distributed and has high utilization potential, and is mostly used for the construction of wind farms. However, due to the very uneven distribution of wind resources, regional wind resource assessment must be carried out before the construction of wind farms to better plan the layout of wind turbines and build a wind power system of appropriate scale, so as to improve the utilization rate of wind resources. Due to the randomness and uncontrollability of wind speed, a large amount of long-term observation data is required to accurately evaluate the wind energy resources in a certain area in order to reduce the uncertainty brought by wind speed changes. However, collecting a large amount of long-term observation data is not only difficult to operate, but also time-consuming. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method, device and equipment for analyzing multi-source satellite offshore regional wind resource maps, which can effectively ensure the integrity of regional data, the consistency with actual observations, and the data accuracy, and can also reduce the difficulty of collecting data for wind resource assessment and prediction.

[0004] In the first aspect, the present invention provides a method for analyzing multi-source satellite offshore regional wind resource maps, including: Preprocessing the multi-source original sea surface wind field in the study sea area to determine the multi-source target sea surface wind field corresponding to the first grid in the study sea area, where the first grid is the grid whose corresponding multi-source original sea surface wind field passes through multi-level wind field quality control; Based on the anemometry station data corresponding to the study sea area, as well as the multi-source target sea surface wind field and reanalysis data corresponding to the first grid, constructing an average sea surface wind field set, where the average sea surface wind field set includes the satellite average sea surface wind field and reanalysis average sea surface wind field corresponding to the first grid at different sea surface heights; Training a machine learning model by using the average sea surface wind field set and the sea surface characteristic data corresponding to the study sea area, so as to predict, through the trained machine learning model, the satellite average sea surface wind field corresponding to the second grid at different sea surface heights based on the reanalysis data corresponding to the second grid in the study sea area, where the second grid is the grid lacking multi-source original sea surface wind field or whose corresponding multi-source original sea surface wind field does not pass through multi-level wind field quality control or the grid with fewer observation times; Generating a wind resource map corresponding to the study sea area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights respectively.

[0005] In one embodiment, preprocessing the multi-source original sea surface wind field of the study sea area to determine the multi-source target sea surface wind field corresponding to the first grid in the study sea area, including: Performing spatio-temporal matching on the offshore buoy data corresponding to the study sea area and the multi-source original sea surface wind field; Using the offshore buoy data to perform multi-level wind field quality control on the multi-source original sea surface wind field to extract the multi-source original sea surface wind field passing through the multi-level wind field quality control; Fitting a wind speed correction coefficient based on the offshore buoy data, and using the wind speed correction coefficient to perform wind speed quality correction on the multi-source original sea surface wind field passing through the multi-level wind field quality control to obtain a multi-source intermediate sea surface wind field; Interpolating the multi-source intermediate sea surface wind field into a grid with a specified resolution to obtain the multi-source target sea surface wind field corresponding to the first grid in the study sea area.

[0006] In one embodiment, constructing an average sea surface wind field set based on the wind measurement station data corresponding to the study sea area, and the multi-source target sea surface wind field and reanalysis data corresponding to the first grid, including: Fitting the wind shear index corresponding to the study sea area at different time periods and different sea surface heights based on the wind measurement station data corresponding to the study sea area; Interpolating the multi-source target sea surface wind field corresponding to the first grid using the wind shear index to obtain the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights; Performing spatio-temporal matching on the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights and the reanalysis data corresponding to the first grid; Based on the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights and its matched reanalysis data, establishing a wind speed Weibull model corresponding to the first grid for generating the average sea surface wind field set.

[0007] In one embodiment, establishing a wind speed Weibull model corresponding to the first grid based on the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights and its matched reanalysis data for generating the average sea surface wind field set, including: Randomly combining the multi-source target sea surface wind fields corresponding to the first grid at different sea surface heights to obtain a sea surface wind field combination corresponding to the first grid at different sea surface heights, where the sea surface wind field combination includes at least two source target sea surface wind fields; Based on the sea surface wind field combination and its matched reanalysis data, establishing a wind speed Weibull model corresponding to the first grid; According to the wind speed probability distribution described by the wind speed Weibull model, calculating the satellite sea surface average wind speed and satellite sea surface average wind direction to obtain the satellite average sea surface wind field, and calculating the reanalysis sea surface average wind speed and reanalysis sea surface average wind direction to obtain the reanalysis average sea surface wind field.

[0008] In one embodiment, training a machine learning model by using an average sea surface wind field set and sea surface characteristic data corresponding to a research sea area includes: For any first grid, using the satellite average sea surface wind field corresponding to the first grid as the learning target, and using the reanalysis average sea surface wind field and sea surface characteristic data corresponding to the first grid and its neighboring first grids as the model inputs to train the machine learning model.

[0009] In one embodiment, predicting the satellite average sea surface wind field corresponding to a second grid at different sea surface heights based on the reanalysis data corresponding to the second grid in the research sea area by using the trained machine learning model includes: Searching for the second grid in the research sea area; For any second grid, determining the reanalysis average sea surface wind field corresponding to the second grid and its neighboring second grids based on the reanalysis data, and inputting the reanalysis average sea surface wind field and sea surface characteristic data corresponding to the second grid and its neighboring second grids into the trained machine learning model to obtain the satellite average sea surface wind field corresponding to the second grid at different sea surface heights.

[0010] In one embodiment, generating a wind resource map corresponding to the research sea area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights includes: Obtaining the satellite average sea surface wind field corresponding to the research sea area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights; Calculating one or more of the effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, and wind direction and wind speed frequency distribution according to the satellite average sea surface wind field corresponding to the research sea area to obtain the wind resource map corresponding to the research sea area.

[0011] In a second aspect, the present invention further provides a multi-source satellite offshore area wind resource map analysis device, including: A preprocessing module, configured to preprocess the multi-source original sea surface wind fields of the research sea area to determine the multi-source target sea surface wind field corresponding to the first grid in the research sea area, where the first grid is the grid whose corresponding multi-source original sea surface wind field passes through multi-level wind field quality control; An average sea surface wind field construction module, configured to construct an average sea surface wind field set based on the anemometer data corresponding to the research sea area, and the multi-source target sea surface wind field and reanalysis data corresponding to the first grid, where the average sea surface wind field set includes the satellite average sea surface wind field and reanalysis average sea surface wind field corresponding to the first grid at different sea surface heights; A missing sea surface wind field prediction module is used to train a machine learning model by using an average sea surface wind field set and sea surface characteristic data corresponding to a research sea area, so as to predict the satellite average sea surface wind field corresponding to different sea surface heights of a second grid in the research sea area through the trained machine learning model, where the second grid is a grid that lacks multi-source original sea surface wind fields or the corresponding multi-source original sea surface wind fields fail to pass multi-level wind field quality control; A wind resource map generation module is used to generate a wind resource map corresponding to the research sea area based on the satellite average sea surface wind fields corresponding to different sea surface heights of the first grid and the second grid respectively.

[0012] In a third aspect, the present invention further provides an electronic device, including a processor and a memory, where the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the method according to any one of the first aspect.

[0013] In a fourth aspect, the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores computer executable instructions, and when the computer executable instructions are called and executed by the processor, the computer executable instructions cause the processor to implement the method according to any one of the first aspect.

[0014] A method, device and equipment for analyzing a multi-source satellite offshore regional wind resource atlas provided by an embodiment of the present invention first preprocesses a multi-source original sea surface wind field in a research sea area to determine a multi-source target sea surface wind field corresponding to a first grid in the research sea area, where the first grid is a grid whose corresponding multi-source original sea surface wind field passes through multi-level wind field quality control; then constructs an average sea surface wind field set based on the anemometer data corresponding to the research sea area, as well as the multi-source target sea surface wind field and reanalysis data corresponding to the first grid, and the average sea surface wind field set includes a satellite average sea surface wind field and a reanalysis average sea surface wind field corresponding to the first grid at different sea surface heights; then uses the average sea surface wind field set and the sea surface characteristic data corresponding to the research sea area to train a machine learning model, so as to, through the trained machine learning model, predict the satellite average sea surface wind field corresponding to a second grid in the research sea area at different sea surface heights based on the reanalysis data corresponding to the second grid in the research sea area, where the second grid is a grid lacking a multi-source original sea surface wind field or whose corresponding multi-source original sea surface wind field fails to pass multi-level wind field quality control; finally, generates a wind resource atlas corresponding to the research sea area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights respectively. The above method trains a machine learning model based on the target sea surface wind field observed by multi-source satellites, reanalysis data and sea surface characteristic data at the first grid, and then uses the machine learning model to calculate the satellite average sea surface wind field in the area without satellite observations or the second grid with missing data, and then generates the wind resource atlas of the entire research sea area. On the one hand, the relatively easily obtained reanalysis data and sea surface characteristic data are used to ensure the integrity, consistency with actual observations and data accuracy of regional data through the machine learning model. On the other hand, it reduces the difficulty of collecting data during wind resource assessment and prediction, reduces the time cost of data collection, and provides a reference basis for reducing the construction risk of wind resource-related projects and improving the benefits of wind energy resource construction.

[0015] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, claims and drawings.

[0016] To make the above objectives, features and advantages of the present invention more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flow chart of a method for analyzing multi-source satellite offshore regional wind resource maps provided by an embodiment of the present invention; Figure 2 It is a technical framework diagram of a method for analyzing multi-source satellite offshore regional wind resource maps provided by an embodiment of the present invention; Figure 3 It is a schematic structural diagram of a device for analyzing multi-source satellite offshore regional wind resource maps provided by an embodiment of the present invention; Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Specific Embodiments

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0020] Currently, the related technologies have problems such as high operation difficulty and long time consumption. Based on this, the embodiments of the present invention provide a method, device, and equipment for analyzing multi-source satellite offshore regional wind resource maps, which can effectively ensure the integrity of regional data, consistency with actual observations, and data accuracy, and can also reduce the difficulty of collecting data for wind resource assessment and prediction. At the same time, it has advantages such as simple algorithms, easy implementation, wide coverage, and strong practicability.

[0021] For the convenience of understanding this embodiment, first, a method for analyzing multi-source satellite offshore regional wind resource maps disclosed in the embodiments of the present invention will be introduced in detail. Refer to Figure 1 The schematic flow chart of a method for analyzing multi-source satellite offshore regional wind resource maps shown, and this method mainly includes the following steps S102 to step S108: Step S102, preprocess the multi-source original sea surface wind field of the research sea area to determine the multi-source target sea surface wind field corresponding to the first grid in the research sea area.

[0022] The first grid is the grid of the corresponding multi-source original sea surface wind field that has passed through multi-level wind field quality control. The preprocessing includes spatio-temporal matching, multi-level wind field quality control, wind field quality correction, and interpolation, etc. In one example, the marine buoy data corresponding to the study sea area can be spatio-temporally matched with the multi-source original sea surface wind field. On this basis, the multi-source original sea surface wind field data is subjected to multi-level wind field quality control and wind field quality correction using the marine buoy data to obtain the multi-source intermediate sea surface wind field. Finally, the intermediate sea surface wind fields from different sources are interpolated into a grid with the same resolution, and the multi-source target sea surface wind field corresponding to the first grid can be obtained.

[0023] Step S104: Based on the wind measurement station data corresponding to the study sea area, as well as the multi-source target sea surface wind field and reanalysis data corresponding to the first grid, construct an average sea surface wind field set.

[0024] Among them, the average sea surface wind field set includes the satellite average sea surface wind field and the reanalysis average sea surface wind field corresponding to the first grid at different sea surface heights. The satellite average sea surface wind field includes the satellite sea surface average wind speed and the satellite sea surface average wind direction. The reanalysis average sea surface wind field includes the reanalysis sea surface average wind speed and the reanalysis sea surface average wind direction. In one example, the wind shear index corresponding to the study sea area at different time periods and different sea surface heights can be fitted using the wind measurement station data; the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights can be obtained by interpolating the multi-source target sea surface wind field using the wind shear index; on this basis, the reanalysis data is spatio-temporally matched, and a wind speed Weibull model describing the probability distribution of wind speed is constructed using the matched data set, and the satellite average sea surface wind field and the reanalysis average sea surface wind field can be calculated respectively based on the probability distribution of wind speed described by the wind speed Weibull model.

[0025] Step S106: Use the average sea surface wind field set and the sea surface characteristic data corresponding to the study sea area to train a machine learning model, so as to predict the satellite average sea surface wind field corresponding to the second grid at different sea surface heights based on the reanalysis data corresponding to the second grid within the study sea area through the trained machine learning model.

[0026] Among them, the sea surface characteristic data includes sea surface height, atmospheric pressure, sea surface temperature, sea surface humidity, etc. The second grid is a grid that lacks the multi-source original sea surface wind field or its corresponding multi-source original sea surface wind field has not passed through multi-level wind field quality control. In one example, taking the satellite average sea surface wind field corresponding to the first grid as the learning target, and taking the reanalysis average sea surface wind field and sea surface characteristic data corresponding to the first grid and its neighborhood as the model input, train a machine learning model (such as XGBoost); for any second grid, input the reanalysis average sea surface wind field and sea surface characteristic data corresponding to the second grid and its neighborhood into the trained machine learning model (such as XGBoost), and the satellite average sea surface wind field corresponding to the second grid can be obtained.

[0027] Step S108: Generate a wind resource atlas for the study area based on the satellite mean sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights.

[0028] Among them, the wind resource atlas can describe the distribution of wind resources such as the effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, and wind direction and wind speed frequency distribution corresponding to each grid in the study area. In one example, based on the satellite mean sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights, the satellite mean sea surface wind field corresponding to the entire study area can be obtained. Based on this satellite mean sea surface wind field, wind resources such as effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, and wind direction and wind speed frequency distribution are calculated to obtain the wind resource atlas.

[0029] The multi-source satellite offshore regional wind resource atlas analysis method provided by the embodiments of the present invention trains a machine learning model based on the target sea surface wind field, reanalysis data, and sea surface characteristic data observed by multi-source satellites at the first grid, and then uses this machine learning model to calculate the satellite mean sea surface wind field in the area without satellite observations or the second grid with missing data, and further generates the wind resource atlas for the entire study area. On the one hand, the relatively easy-to-obtain reanalysis data and sea surface characteristic data are used to ensure the integrity, consistency with actual observations, and data accuracy of regional data through the machine learning model. On the other hand, it reduces the difficulty of collecting data during wind resource assessment and prediction, reduces the time cost of data collection, and provides a reference basis for reducing the construction risk of wind resource-related projects and improving the efficiency of wind energy resource construction.

[0030] For ease of understanding, the embodiments of the present invention provide a specific implementation manner of the multi-source satellite offshore regional wind resource atlas analysis method. See Figure 2The technical framework diagram of a multi-source satellite offshore regional wind resource atlas analysis method shown in the figure, the main technical process of this method is: collect long-term series SAR satellite data (that is, multi-source original sea surface wind field), ERA5 reanalysis data, wind tower data and other sea surface feature data, first pre-process the SAR satellite data mainly including multi-level quality control and wind speed correction; then fit the wind shear index according to the wind tower data, and perform spatiotemporal matching between the SAR satellite data and the ERA5 reanalysis data, establish the wind speed Weibull distribution based on the matching data set, and calculate the observed average wind speed and average wind direction; establish an XGBoost wind field prediction model based on the observed average wind speed and average wind direction, as well as the matching data set and sea surface feature data, and use the model to predict the average wind field in the data missing area to obtain the average wind field of the entire study sea area; finally, based on the average wind field of the entire study sea area, evaluate the wind energy resources from the perspectives of effective wind speed, wind speed Weibull distribution analysis, effective wind energy density distribution, wind energy variation coefficient, wind direction frequency statistics, etc.

[0031] The specific implementation process is as follows: (I) Data collection: Collect long-term series of SAR satellite data in the study area, including original sea surface wind fields observed by multiple-source satellites such as Sentinel-1A, Sentinel-1B, GF3, and 1mCSAR; collect long-term series of ERA5 reanalysis data in the study area; collect offshore buoy data and wind tower data in the study area; collect sea surface characteristic data such as sea surface height, atmospheric pressure, sea surface temperature, and sea surface humidity in the study area.

[0032] (II) Preprocessing of the original sea surface wind field, including the following (2.1) to (2.4): (2.1) Perform spatiotemporal matching of the offshore buoy data and the multi-source original sea surface wind field corresponding to the study area. In one example, assuming that the offshore buoy data is the true wind speed and direction, the multi-source original sea surface wind field is spatiotemporally matched with the offshore buoy data, with a matching time window of 30 minutes and a spatial window of 50 km.

[0033] (2.2) Using offshore buoy data, multi-level wind field quality control is performed on the multi-source original sea surface wind field to extract the multi-source original sea surface wind field that has passed the multi-level wind field quality control.

[0034] Among them, multi-level quality control includes data quality screening and anomaly detection quality control through the quality labels of various products. Not all wind field data need to be involved in the calculation. When the deviation of a single data exceeds 3 times the mean absolute error of the overall data, this data can be considered an anomaly that needs to be eliminated.

[0035] In one example, the wind direction error can be calculated as follows: ; In the formula, represents the wind direction value of the th matching point of the original sea surface wind field, represents the buoy wind direction value of the th matching point, is the number of matching points.

[0036] In one example, the wind speed error can be calculated according to the following formula: ; In the formula, represents the wind speed value of the th matching point of the original sea surface wind field, represents the buoy wind speed value of the th matching point, is the number of matching points.

[0037] Based on the above wind direction error and wind speed error, the mean absolute wind direction error and the mean absolute wind speed error can be determined respectively, which are used for multi-level wind field quality control.

[0038] (2.3) Fitting the wind speed correction coefficient based on the offshore buoy data, and using the wind speed correction coefficient to perform wind speed quality correction on the multi-source original sea surface wind field passing through multi-level wind field quality control to obtain a multi-source intermediate sea surface wind field. In one example, the sea surface wind speed in the multi-source original sea surface wind field is corrected according to the spatio-temporal matching data result. Mainly, linear fitting is performed based on the matched sea surface wind speed, and the fitting correction parameters for each satellite are respectively , , , , , , , . The corrected sea surface wind speed is calculated according to the correction parameter as follows: ; ; ; ; Among them, is the corrected sea surface wind speed of Sentinel-1A, is the sea surface wind speed before correction of Sentinel-1A, is the corrected sea surface wind speed of Sentinel-1B, is the sea surface wind speed before correction of Sentinel-1B, is the corrected sea surface wind speed of GF3, is the sea surface wind speed before calibration for GF3, is the sea surface wind speed after calibration for 1m CSAR, is the sea surface wind speed before calibration for 1m CSAR.

[0039] (2.4) Interpolate the multi-source intermediate sea surface wind field into a grid with a specified resolution to obtain the multi-source target sea surface wind field corresponding to the first grid in the study area. In practical applications, due to the different spatial resolutions of different SAR data, it is also necessary to interpolate the calibrated multi-source intermediate sea surface wind field into a grid with a unified resolution.

[0040] (III) Fitting the wind shear index: Based on the wind measurement station data corresponding to the study area, fit the wind shear index corresponding to different time periods and different sea surface heights in the study area.

[0041] Among them, the wind shear index is used to measure the characteristics of the wind speed changing with the vertical height, and is related to the sea surface height, surface characteristics, weather and climate, and seasonal changes. The wind shear index can be calculated using the exponential law method, and the expression is: ; Among them, and are the average sea surface wind speeds at heights and respectively, is the wind shear index. Since the near-surface upper and lower layer turbulent exchange effects are different during the day and night, there are also differences between different height layers. The wind shear indices for day and night and low, middle, and high layers are respectively fitted using the wind measurement tower data of different height layers.

[0042] (IV) Wind field data interpolation, including the following (4.1) to (4.3): (4.1) Interpolate the multi-source target sea surface wind field corresponding to the first grid using the wind shear index to obtain the multi-source target sea surface wind field corresponding to different sea surface heights in the first grid. In an example, for any sea surface height to be interpolated (such as any one of the low, middle, and high layers), the sea surface wind speed included in the target sea surface wind field and the sea surface height at which it is located are known, and the wind shear index between the sea surface height to be interpolated and the sea surface height at which the target sea surface wind field is located is also known. Using the formula in (III), the target sea surface wind speed corresponding to the sea surface height to be interpolated can be obtained, and then the target sea surface wind field corresponding to the sea surface height to be interpolated can be obtained.

[0043] (4.2) Perform spatio-temporal matching on the multi-source target sea surface wind field corresponding to different sea surface heights in the first grid and the reanalysis data corresponding to the first grid. In an example, the ERA5 reanalysis data is interpolated to the same resolution according to the observation time and observation longitude and latitude data of the multi-source target sea surface wind field to obtain the matched ERA5 reanalysis data.

[0044] (4.3)Based on the multi-source target sea surface wind fields corresponding to different sea surface heights in the first grid and their matched reanalysis data, establish a Weibull model of wind speed corresponding to the first grid to generate a set of mean sea surface wind fields.

[0045] In a specific implementation manner, first, randomly combine the multi-source target sea surface wind fields corresponding to different sea surface heights in the first grid to obtain a sea surface wind field combination corresponding to different sea surface heights in the first grid. The sea surface wind field combination includes at least two-source target sea surface wind fields; then, based on the sea surface wind field combination and its matched reanalysis data, establish a Weibull model of wind speed corresponding to the first grid; finally, according to the wind speed probability distribution described by the Weibull model of wind speed, calculate the satellite sea surface average wind speed and satellite sea surface average wind direction to obtain the satellite average sea surface wind field, and calculate the reanalysis sea surface average wind speed and reanalysis sea surface average wind direction to obtain the reanalysis average sea surface wind field.

[0046] (5)Construction of the XGBoost wind field prediction model: For any first grid, use the satellite average sea surface wind field corresponding to the first grid as the learning target, and use the reanalysis average sea surface wind field and sea surface characteristic data corresponding to the first grid and its neighboring first grids as the model inputs to train the machine learning model. Specifically: (5.1)Sample data generation: Convert the satellite average sea surface wind field and the reanalysis average sea surface wind field into the UV components of the wind field, which are U_mean_sar, V_mean_sar, U_mean_era5, and V_mean_era5 respectively; generate the training samples of XGBoost according to the average UV components of the wind field and the data of atmospheric pressure, temperature, and humidity. Using U_mean_sar and V_mean_sar as the learning targets, the U_mean_era5 and V_mean_era5 data of the grid points of U_mean_sar and V_mean_sar and their surrounding n*n (n = 1, 3, 5, 7...) grids, as well as the atmospheric pressure, sea surface temperature, and sea surface humidity of the grids, are used as training features to obtain a sample data set, in which 80% is randomly selected as the training set and 20% as the validation set.

[0047] (5.2) Set different surrounding n*n grids, where n ranges from 3 to 21. The K-fold cross-validation method is used to adjust the XGBoost parameters, and the best model parameters and the best n value are selected. K-fold cross-validation divides the sample dataset into K parts, with one part as the validation set and the remaining K-1 parts as the training set. Each time, the model is trained with the training set, and the model error is calculated by testing the model with the validation set. The cross-validation is repeated K times, and the K model validation results are returned. The average of the K results is used as the final error of the model. Here, K = 10 is set. The performance of the wind field prediction model is evaluated through various error statistical methods, and the error statistical methods are the correlation coefficient (R), standard deviation (STD), mean absolute error (MAE), root mean square error (RMSE), and central root mean square error (E’).

[0048] ; ; ; ; .

[0049] (5.3) Construct the XGBoost model parameter space. The XGBoost model is an additive training model based on the Boosting integration idea. It uses the forward distribution algorithm for greedy learning. Each iteration learns a CART tree to fit the prediction results of the previous k-1 trees and the residuals of the sample values. The final prediction value is the sum of the prediction values of each decision tree: = .

[0050] The main parameters of the XGBoost model mainly include the depth of the tree structure max_depth, the minimum weight threshold of the child node min_child_weight. If the weight of a tree node is less than this threshold, the subtree will not be split anymore. The loss reduction threshold gamma for the decision tree splitting of XGBoost. If the model shows overfitting, the subsampling parameter subsample, the feature sampling ratio of the entire tree colsample_bytree, the feature sampling ratio of a certain layer colsample_bylevel, the feature sampling ratio of a certain tree node colsample_bynode, and the L1L2 regularization parameter terms reg_alpha, reg_lambda need to be adjusted. By adjusting the XGBoost parameters, the final XGBoost model, that is, the average wind field prediction model, is obtained.

[0051] (6) Average wind field calculation: Search for the second grid within the study area. For any second grid, determine the reanalysis average sea surface wind field corresponding to this second grid and its neighboring second grids based on the reanalysis data. Input the reanalysis average sea surface wind field corresponding to this second grid and its neighboring second grids and the sea surface characteristic data into the trained machine learning model to obtain the satellite average sea surface wind field corresponding to the second grid at different sea surface heights.

[0052] Specifically: Since the sea surface wind field has undergone multiple levels of quality control and there are many data gaps, search for the positions of the grid points with data gaps and the corresponding sea surface characteristic data such as ERA5 reanalysis data, atmospheric pressure, sea surface temperature, and sea surface humidity at these positions and their surrounding grid points. Use the XGBoost wind field prediction model constructed in the previous step to calculate the satellite average sea surface wind field of the grid points with data gaps, and obtain the satellite average sea surface wind field of the complete area for regional wind resource assessment.

[0053] (7) Wind resource atlas calculation and assessment: First, based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights respectively, obtain the satellite average sea surface wind field corresponding to the study area; then, according to the satellite average sea surface wind field corresponding to the study area, calculate one or more of the effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, and wind direction and wind speed frequency distribution to obtain the wind resource atlas corresponding to the study area.

[0054] Specifically: Calculate the effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, and wind direction and wind speed frequency distribution according to the satellite average sea surface wind field obtained in the previous step. Evaluate the wind energy magnitude according to the effective wind energy density, evaluate the wind energy stability according to the wind energy variation coefficient, and obtain the dominant wind direction and the change in wind direction stability in the target area through the wind direction and wind speed frequency distribution, so as to comprehensively evaluate the richness and stability of the regional wind resources.

[0055] In summary, as a common clean energy, wind energy has great development potential, especially in the field of wind power generation. However, the assessment of wind energy resources often requires years or even decades of observational data, which is difficult to collect in actual construction applications, and the data integrity is low. To address this defect, the embodiments of the present invention utilize the relatively easily obtainable ERA5 reanalysis data, fully consider the sea surface characteristic data related to the wind field and the relationship between the wind unit and its surrounding units, and ensure the integrity, rationality, and consistency of the regional data by establishing an XGBoost wind field prediction model, reducing the difficulty of collecting data during wind resource assessment and prediction, reducing the time cost of data collection, and providing a reference basis for reducing the construction risk of wind resource-related projects and improving the efficiency and utilization rate of wind energy resource construction.

[0056] Based on the foregoing embodiments, an embodiment of the present invention provides a multi-source satellite offshore regional wind resource atlas analysis device. Refer to Figure 3 the structural schematic diagram of a multi-source satellite offshore regional wind resource atlas analysis device shown in The preprocessing module 302 is configured to preprocess the multi-source original sea surface wind field of the research sea area to determine the multi-source target sea surface wind field corresponding to the first grid in the research sea area. The first grid is a grid whose corresponding multi-source original sea surface wind field passes through multi-level wind field quality control; The mean sea surface wind field construction module 304 is configured to construct a set of mean sea surface wind fields based on the anemometer data corresponding to the research sea area, the multi-source target sea surface wind field corresponding to the first grid, and the reanalysis data. The set of mean sea surface wind fields includes the satellite mean sea surface wind field and the reanalysis mean sea surface wind field corresponding to the first grid at different sea surface heights; The missing sea surface wind field prediction module 306 is configured to train a machine learning model by using the set of mean sea surface wind fields and the sea surface characteristic data corresponding to the research sea area, so as to, based on the reanalysis data corresponding to the second grid in the research sea area, predict the satellite mean sea surface wind field corresponding to the second grid at different sea surface heights through the trained machine learning model. The second grid is a grid that lacks multi-source original sea surface wind field or whose corresponding multi-source original sea surface wind field fails to pass multi-level wind field quality control; The wind resource atlas generation module 308 is configured to generate a wind resource atlas corresponding to the research sea area based on the satellite mean sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights respectively.

[0057] The multi-source satellite offshore regional wind resource atlas analysis device provided by the embodiment of the present invention trains a machine learning model based on the target sea surface wind field observed by multi-source satellites, reanalysis data, and sea surface characteristic data at the first grid, and then uses the machine learning model to calculate the satellite mean sea surface wind field in the area without satellite observations or the second grid with missing data, and further generates a wind resource atlas for the entire research sea area. On the one hand, the relatively easily obtained reanalysis data and sea surface characteristic data are used to ensure the integrity, consistency with actual observations, and data accuracy of regional data through the machine learning model. On the other hand, it reduces the difficulty of collecting data during wind resource assessment and prediction, reduces the time cost of data collection, and provides a reference basis for reducing the construction risk of wind resource-related projects and improving the efficiency of wind energy resource construction.

[0058] In an implementation manner, the preprocessing module 302 is specifically configured to: Perform spatio-temporal matching on the offshore buoy data corresponding to the research sea area and the multi-source original sea surface wind field; Using buoy data at sea, perform multi-level quality control on multi-source original sea surface wind fields to extract multi-source original sea surface wind fields that pass the multi-level quality control. Fit a wind speed correction coefficient based on buoy data at sea, and use the wind speed correction coefficient to perform wind speed quality correction on the multi-source original sea surface wind fields that pass the multi-level quality control to obtain multi-source intermediate sea surface wind fields. Interpolate the multi-source intermediate sea surface wind fields into a grid with a specified resolution to obtain the multi-source target sea surface wind fields corresponding to the first grid in the study area.

[0059] In one implementation, the average sea surface wind field construction module 304 is specifically configured to: Based on the anemometer data corresponding to the study area, fit the wind shear index corresponding to different time periods and different sea surface heights in the study area. Use the wind shear index to interpolate the multi-source target sea surface wind fields corresponding to the first grid to obtain the multi-source target sea surface wind fields corresponding to the first grid at different sea surface heights. Perform spatio-temporal matching on the multi-source target sea surface wind fields corresponding to the first grid at different sea surface heights and the reanalysis data corresponding to the first grid. Based on the multi-source target sea surface wind fields corresponding to the first grid at different sea surface heights and their matched reanalysis data, establish a wind speed Weibull model corresponding to the first grid to generate a set of average sea surface wind fields.

[0060] In one implementation, the average sea surface wind field construction module 304 is specifically configured to: Randomly combine the multi-source target sea surface wind fields corresponding to the first grid at different sea surface heights to obtain a sea surface wind field combination corresponding to the first grid at different sea surface heights, where the sea surface wind field combination includes at least two source target sea surface wind fields. Based on the sea surface wind field combination and its matched reanalysis data, establish a wind speed Weibull model corresponding to the first grid. According to the wind speed probability distribution described by the wind speed Weibull model, calculate the satellite sea surface average wind speed and satellite sea surface average wind direction to obtain the satellite average sea surface wind field, and calculate the reanalysis sea surface average wind speed and reanalysis sea surface average wind direction to obtain the reanalysis average sea surface wind field.

[0061] In one implementation, the missing sea surface wind field prediction module 306 is specifically configured to: For any first grid, use the satellite average sea surface wind field corresponding to the first grid as the learning target, and use the reanalysis average sea surface wind field and sea surface feature data corresponding to the first grid and its neighboring first grids as model inputs to train the machine learning model.

[0062] In one implementation, the missing sea surface wind field prediction module 306 is specifically configured to: Search for the second grid within the study area; For any second grid, determine the reanalysis mean sea surface wind field corresponding to the second grid and its neighboring second grids based on the reanalysis data, and input the reanalysis mean sea surface wind field corresponding to the second grid and its neighboring second grids and the sea surface characteristic data into the trained machine learning model to obtain the satellite mean sea surface wind field corresponding to the second grid at different sea surface heights.

[0063] In one implementation, the wind resource map generation module 308 is specifically configured to: Based on the satellite mean sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights, obtain the satellite mean sea surface wind field corresponding to the study area; According to the satellite mean sea surface wind field corresponding to the study area, calculate one or more of the effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, and wind direction and wind speed frequency distribution to obtain the wind resource map corresponding to the study area.

[0064] The device provided by the embodiments of the present invention has the same implementation principle and the same technical effects as the foregoing method embodiments. For the sake of brief description, for the parts not mentioned in the device embodiments, reference may be made to the corresponding contents in the foregoing method embodiments.

[0065] The embodiments of the present invention provide an electronic device. Specifically, the electronic device includes a processor and a storage device; a computer program is stored on the storage device, and the computer program executes the method according to any one of the foregoing implementation manners when being run by the processor.

[0066] Figure 4 FIG. 19 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device 100 includes: a processor 40, a memory 41, a bus 42, and a communication interface 43. The processor 40, the communication interface 43, and the memory 41 are connected through the bus 42; the processor 40 is configured to execute an executable module stored in the memory 41, such as a computer program.

[0067] Among them, the memory 41 may include a high-speed random access memory (RAM, Random Access Memory), and may also include a non-volatile memory, such as at least one disk memory. Through at least one communication interface 43 (which may be wired or wireless), a communication connection between the system network element and at least one other network element is realized, and the Internet, wide area network, local area network, metropolitan area network, etc. can be used.

[0068] The bus 42 may be an ISA bus, a PCI bus, an EISA bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 4It is represented by only one bidirectional arrow, but it does not mean that there is only one bus or one type of bus.

[0069] Among them, the memory 41 is used to store a program. After receiving an execution instruction, the processor 40 executes the program. The method executed by the device defined by the flow process disclosed in any embodiment of the foregoing embodiments of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0070] The processor 40 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor 40 or by instructions in the form of software. The above-mentioned processor 40 may be a general-purpose processor, including a central processing unit (Central Processing Unit, abbreviated as CPU), a network processor (Network Processor, abbreviated as NP), etc.; it may also be a digital signal processor (Digital Signal Processing, abbreviated as DSP), an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), a field-programmable gate array (Field-Programmable Gate Array, abbreviated as FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and combines its hardware to complete the steps of the above method.

[0071] The computer program product of the readable storage medium provided by the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the foregoing method embodiments. For the specific implementation, reference can be made to the foregoing method embodiments and will not be elaborated herein.

[0072] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0073] Finally, it should be noted that the above-mentioned embodiments are only specific embodiments of the present invention, used to illustrate the technical solution of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions described in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes, or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A multi-source satellite offshore regional wind resource atlas analysis method, characterized in that Including: Preprocessing the multi-source original sea surface wind field of the study area to determine the multi-source target sea surface wind field corresponding to the first grid in the study area, where the first grid is the grid whose corresponding multi-source original sea surface wind field passes through multi-level wind field quality control; Constructing an average sea surface wind field set based on the anemometry station data corresponding to the study area, the multi-source target sea surface wind field corresponding to the first grid, and the reanalysis data, where the average sea surface wind field set includes the satellite average sea surface wind field and the reanalysis average sea surface wind field corresponding to the first grid at different sea surface heights; Training a machine learning model using the average sea surface wind field set and the sea surface characteristic data corresponding to the study area, so as to predict, through the trained machine learning model, the satellite average sea surface wind field corresponding to the second grid in the study area at different sea surface heights based on the reanalysis data corresponding to the second grid in the study area, where the second grid is the grid lacking the multi-source original sea surface wind field or whose corresponding multi-source original sea surface wind field fails to pass the multi-level wind field quality control; Generating a wind resource map corresponding to the study area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights.

2. The multi-source satellite offshore regional wind resource atlas analysis method according to claim 1, characterized in that Preprocessing the multi-source original sea surface wind field of the study area to determine the multi-source target sea surface wind field corresponding to the first grid in the study area, including: Performing spatio-temporal matching on the offshore buoy data corresponding to the study area and the multi-source original sea surface wind field; Using the offshore buoy data to perform multi-level wind field quality control on the multi-source original sea surface wind field to extract the multi-source original sea surface wind field passing through the multi-level wind field quality control; Fitting a wind speed correction coefficient based on the offshore buoy data, and using the wind speed correction coefficient to perform wind speed quality correction on the multi-source original sea surface wind field passing through the multi-level wind field quality control to obtain a multi-source intermediate sea surface wind field; Interpolating the multi-source intermediate sea surface wind field into a grid with a specified resolution to obtain the multi-source target sea surface wind field corresponding to the first grid in the study area.

3. The multi-source satellite offshore regional wind resource atlas analysis method according to claim 1, wherein Constructing an average sea surface wind field set based on the anemometry station data corresponding to the study area, the multi-source target sea surface wind field corresponding to the first grid, and the reanalysis data, including: Fitting the wind shear index corresponding to the study area at different time periods and different sea surface heights based on the anemometry station data corresponding to the study area; Interpolating the multi-source target sea surface wind field corresponding to the first grid using the wind shear index to obtain the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights; Performing spatio-temporal matching on the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights and the reanalysis data corresponding to the first grid; Establishing a wind speed Weibull model corresponding to the first grid based on the multi-source target sea surface wind field corresponding to the first grid at different sea surface heights and its matched reanalysis data for generating the average sea surface wind field set.

4. The multi-source satellite offshore regional wind resource atlas analysis method according to claim 3, characterized in that Based on the multi-source target sea surface wind fields corresponding to different sea surface heights in the first grid and the corresponding reanalysis data, establish the wind speed Weibull model corresponding to the first grid to generate the average sea surface wind field set, including: Randomly combine the multi-source target sea surface wind fields corresponding to different sea surface heights in the first grid to obtain the sea surface wind field combinations corresponding to different sea surface heights in the first grid, where the sea surface wind field combinations include at least two-source target sea surface wind fields; Based on the sea surface wind field combinations and the corresponding reanalysis data, establish the wind speed Weibull model corresponding to the first grid; According to the wind speed probability distribution described by the wind speed Weibull model, calculate the satellite sea surface average wind speed and satellite sea surface average wind direction to obtain the satellite average sea surface wind field, and calculate the reanalysis sea surface average wind speed and reanalysis sea surface average wind direction to obtain the reanalysis average sea surface wind field.

5. The multi-source satellite offshore regional wind resource atlas analysis method according to claim 1, characterized in that Use the average sea surface wind field set and the sea surface characteristic data corresponding to the study area to train the machine learning model, including: For any one of the first grids, use the satellite average sea surface wind field corresponding to the first grid as the learning target, and use the reanalysis average sea surface wind field and sea surface characteristic data corresponding to the first grid and its neighboring first grids as the model inputs to train the machine learning model.

6. The multi-source satellite-based offshore regional wind resource atlas analysis method according to claim 1, characterized in that Through the trained machine learning model, based on the reanalysis data corresponding to the second grid in the study area, predict the satellite average sea surface wind fields corresponding to different sea surface heights of the second grid, including: Search for the second grid in the study area; For any one of the second grids, determine the reanalysis average sea surface wind fields corresponding to the second grid and its neighboring second grids based on the reanalysis data, and input the reanalysis average sea surface wind fields and sea surface characteristic data corresponding to the second grid and its neighboring second grids into the trained machine learning model to obtain the satellite average sea surface wind fields corresponding to different sea surface heights of the second grid.

7. The multi-source satellite offshore regional wind resource atlas analysis method according to claim 1, characterized in that Based on the satellite average sea surface wind fields corresponding to different sea surface heights of the first grid and the second grid respectively, generate the wind resource map corresponding to the study area, including: Based on the satellite average sea surface wind fields corresponding to different sea surface heights of the first grid and the second grid respectively, obtain the satellite average sea surface wind field corresponding to the study area; According to the satellite average sea surface wind field corresponding to the study area, calculate one or more of the effective wind energy density, maximum wind speed, effective wind hours, wind energy variation coefficient, wind direction and wind speed frequency distribution to obtain the wind resource map corresponding to the study area.

8. A multi-source satellite offshore wind resource atlas analysis device, characterized in that, Including: A preprocessing module for preprocessing the multi-source original sea surface wind fields in the study area to determine the multi-source target sea surface wind fields corresponding to the first grid in the study area, where the first grid is the grid whose corresponding multi-source original sea surface wind fields pass through multi-level wind field quality control; An average sea surface wind field construction module, configured to construct an average sea surface wind field set based on the wind measurement station data corresponding to the study sea area, and the multi-source target sea surface wind field and reanalysis data corresponding to the first grid. The average sea surface wind field set includes the satellite average sea surface wind field and the reanalysis average sea surface wind field corresponding to the first grid at different sea surface heights; A missing sea surface wind field prediction module, configured to train a machine learning model by using the average sea surface wind field set and the sea surface characteristic data corresponding to the study sea area, so as to predict, through the trained machine learning model, the satellite average sea surface wind field corresponding to the second grid in the study sea area at different sea surface heights based on the reanalysis data corresponding to the second grid in the study sea area. The second grid is a grid that lacks the multi-source original sea surface wind field or the corresponding multi-source original sea surface wind field fails to pass the multi-level wind field quality control or has a small number of observations; A wind resource atlas generation module, configured to generate a wind resource atlas corresponding to the study sea area based on the satellite average sea surface wind fields corresponding to the first grid and the second grid at different sea surface heights respectively; 9. An electronic device, characterized in that, It includes a processor and a memory. The memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the method according to any one of claims 1 to 7; 10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer executable instructions, and when the computer executable instructions are called and executed by the processor, the computer executable instructions cause the processor to implement the method according to any one of claims 1 to 7.

Images