High-altitude ice crystal water content estimation method based on airborne platform

Abstract

The invention provides a high-altitude ice crystal water content estimation method based on an airborne platform, which belongs to the technical field of airborne meteorological radar detection, and specifically comprises the following steps of: constructing echo inversion data of an airborne meteorological radar based on the airborne platform, and abandoning an idealized model for estimating water content based on equally spaced data in the existing method; the characteristic quantity capable of representing the beam width change is provided, and the attention of the classification model on the change relation between the beam width and the high-altitude ice crystal water content is strengthened; a double stratified sampling method is provided, the problem that the number of high-altitude ice crystal samples with different grades and concentrations is unbalanced under the actual condition is solved, the dependence of the high-altitude ice crystal water content on the temperature is reduced from the sample level, robust high-altitude ice crystal water content estimation under an airborne platform is achieved, and the accuracy of high-altitude ice crystal water content estimation is improved. The objective of the invention is to improve the accuracy of high-altitude ice crystal danger level prediction of a subsequent airborne meteorological radar in a cruising stage and guarantee the flight safety of an aircraft.

Description

Technical Field

[0001] The present application relates to the field of airborne meteorological radar detection, and in particular to a method for estimating the water content of high-altitude ice crystals based on an airborne platform. Background Art

[0002] High ice water content (HIWC) is one of the main meteorological threats to aircraft flying at high altitudes. When an aircraft passes through clouds containing a large amount of supercooled water, the supercooled water condenses into ice crystals by hitting the aircraft surface. Icing can cause varying degrees of damage to the aircraft, such as causing the aircraft to stall, deteriorate the aircraft's aerodynamic performance, increase flight fuel consumption, reduce engine performance, and block the pitot tube.

[0003] At present, the sensors that can be used for high-altitude ice crystal detection on aircraft are mainly divided into two categories, namely in-situ detection sensors and meteorological radars. Among them, meteorological radar, as a commonly used airborne meteorological detection equipment, has attracted more and more attention from researchers for its detection performance of high-altitude ice crystals. The high-altitude ice crystal detection principle of airborne meteorological radar is based on the Z-IWC relationship. Generally, the higher the IWC of the ice crystal water content, the larger the Z value, and the greater the corresponding radar echo intensity. Therefore, airborne meteorological radar can achieve indirect estimation of the water content of high-altitude ice crystals and judgment of the danger level by obtaining the intensity information of the echo.

[0004] Existing H IWC estimation usually divides the three-dimensional information of meteorological echoes into evenly spaced grid points, and then estimates the water content of each grid point. Common estimation methods include constructing the relationship equation between reflectivity and water content (Z-IWC). However, these studies are all based on the prediction of water content at evenly divided meteorological grid points. But in actual flight, the volume of meteorological targets detected by radar will change as the beam coverage increases. Obviously, as the beam illumination volume widens with increasing distance, the types of particles within the beam coverage tend to be more complex, and when the size of the meteorological target is smaller than the beam coverage, the wrong estimation of the size of the meteorological target will also affect the accuracy of ice crystal water content prediction. Therefore, in order to further explore the performance of ice crystal water content estimation methods for airborne meteorological radars, it is urgent to carry out research on airborne meteorological radar IWC estimation algorithms on airborne platforms that integrate the effect of increasing beam coverage. Summary of the invention

[0005] In view of this, the present application provides a method for estimating the moisture content of high-altitude ice crystals based on an airborne platform, which solves the problems in the prior art, proposes a beam width characterization operator, and combines features such as radar reflectivity factor to realize a robust method for estimating the moisture content of high-altitude ice crystals under multiple temperature layers, effectively improving the perception capability of airborne meteorological radars to high-altitude ice crystals.

[0006] The present application provides a method for estimating the water content of high-altitude ice crystals based on an airborne platform, which adopts the following technical solutions:

[0007] A method for estimating the water content of high-altitude ice crystals based on an airborne platform comprises the following steps:

[0008] Obtain high-altitude ice crystal data sets as meteorological targets and set the three-dimensional position information of the meteorological targets;

[0009] Establish an airborne weather radar echo inversion model, simulate the real detection scenario of the airborne weather radar, set multiple aircraft initial positions and flight paths according to the location of the weather target, set the aircraft avionics information, and set the airborne weather radar parameter information;

[0010] Extract the characteristics of high-altitude ice crystals covered by the airborne weather radar beam during the flight process and obtain the airborne weather radar echo inversion data;

[0011] A feature matrix for estimating the water content of high-altitude ice crystals for airborne platforms is established. The minimum feature set of the matrix is ​​the radar reflectivity factor Z, temperature T, and beam width characterization quantity W. The radar reflectivity factor Z and temperature T are obtained from the airborne meteorological radar echo inversion data, and the beam width characterization quantity W is the volume between adjacent range gates.

[0012] Construct a label matrix: determine the thresholds of different danger levels of high-altitude ice crystals, quantify the ice crystal water content IWC value in the airborne meteorological radar echo inversion data according to the specified threshold to obtain the high-altitude ice crystal sample label, and establish a data set D, which includes the reflectivity factor Z, temperature T, beam width representation W and the corresponding high-altitude ice crystal sample label Y;

[0013] Perform the first stratified sampling on the dataset D according to the sample label Y to obtain samples of various types with a balanced number;

[0014] In the process of extracting each type of sample, the second stratified sampling is performed according to the temperature T information in the feature set to obtain a data set F with a balanced number of samples in different temperature layers;

[0015] Establish a recognition model, use the data set F as the input of the recognition model for training, and obtain a high-altitude ice crystal water content estimation model;

[0016] The actual acquired high-altitude ice crystal data of the airborne platform is input into the high-altitude ice crystal moisture content estimation model for online testing, and the high-altitude ice crystal moisture content estimation model after the test is used to estimate the high-altitude ice crystal moisture content of the airborne platform.

[0017] Optionally, the method for obtaining the echo inversion data of the airborne weather radar includes: the echo inversion data corresponding to the kth range gate of the radar beam at the current carrier position is the average value of the meteorological data covered in the k-1th to kth range gates, completing the azimuth and elevation scanning and the traversal of different carrier positions to obtain the echo inversion data of the airborne weather radar.

[0018] Optionally, the beam width characterization quantity W=l×h×w, where l is the length corresponding to the range gate; h is the height of the range gate, that is, the Rth n The pitch length corresponding to a range gate; w is the width of the range gate.

[0019] Optionally, setting the aircraft avionics information includes setting the aircraft's flight speed, flight altitude, and heading angle.

[0020] Optionally, setting the airborne weather radar parameter information includes setting the elevation and azimuth beam widths, the elevation and azimuth scanning ranges, the maximum detection range, and the actual distance corresponding to a range gate.

[0021] Optionally, the step of determining thresholds for different danger levels of high-altitude ice crystals includes: when IWC <Ag / m 3 For low concentration, Ag / m 3 ≤IWC 3 Medium concentration, IWC ≥ B g / m 3 is a high concentration, where IWC is the ice crystal water content, where A represents the first threshold, B represents the second threshold, and B is greater than A.

[0022] Optionally, the step of establishing the data set D includes:

[0023] The obtained ice crystal water content IWC is quantified according to a given threshold. The specific quantification method is as follows:

[0024]

[0025] The number of samples in the data set is K, and the data set D is specifically expressed as:

[0026]

[0027] Where, Y = [y1 y2…y K ] T is the quantized label vector.

[0028] In summary, this application includes the following beneficial technical effects:

[0029] ​The present application fully considers the problem of inaccurate estimation of ice crystal moisture content level due to beam broadening and increased beam coverage in actual applications of airborne meteorological radar, constructs echo inversion data based on airborne meteorological radar on an airborne platform, abandons the idealized model of moisture content estimation based on equally spaced data in existing methods, proposes characteristic quantities that can characterize changes in beam width, and strengthens the classification model's attention to the relationship between beam width and changes in moisture content of high-altitude ice crystals; proposes a double stratified sampling method to solve the problem of unbalanced number of high-altitude ice crystal samples of different concentrations under actual circumstances, and reduces the dependence of moisture content of high-altitude ice crystals on temperature at the sample level, thereby achieving robust estimation of moisture content of high-altitude ice crystals on an airborne platform, aiming to improve the accuracy of subsequent airborne meteorological radar predictions of high-altitude ice crystal hazard levels during the cruise phase, and ensure the flight safety of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 A flowchart for constructing a high-altitude ice crystal dataset and a label matrix in an embodiment of the present application;

[0032] Figure 2 This is a schematic diagram of the airborne weather radar echo inversion process in the embodiment of the present application;

[0033] Figure 3 Schematic diagram of the principle of the dual data feature sampling method in the embodiment of the present application;

[0034] Figure 4 Schematic diagram of model training and testing of the ice crystal water content estimation model in the embodiment of the present application. DETAILED DESCRIPTION

[0035] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0036] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0037] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0038] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. The illustrations only show the components related to the present application rather than being drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0039] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0040] The embodiment of the present application provides a method for estimating the water content of high-altitude ice crystals based on an airborne platform.

[0041] A method for estimating the water content of high-altitude ice crystals based on an airborne platform comprises the following steps:

[0042] Obtain high-altitude ice crystal datasets as meteorological targets and set the three-dimensional position information of the meteorological targets.

[0043] Establish an airborne weather radar echo inversion model, simulate the real detection scene of airborne weather radar, set multiple initial positions and flight paths of carrier aircraft according to the location of weather targets, set aircraft avionics information, and set airborne weather radar parameter information. Among them, setting aircraft avionics information includes setting the aircraft's flight speed, flight altitude, and heading angle; setting airborne weather radar parameter information includes setting the pitch and azimuth beam width, pitch and azimuth scanning range, maximum detection range, and the actual distance corresponding to a range gate.

[0044] The characteristics of high-altitude ice crystals covered by the airborne meteorological radar beam during the flight process are extracted to obtain the airborne meteorological radar echo inversion data.

[0045] A feature matrix for estimating the water content of high-altitude ice crystals for airborne platforms is established. The minimum feature set of the matrix is ​​radar reflectivity factor Z, temperature T and beam width characterization value W, where radar reflectivity factor Z and temperature T are obtained from airborne meteorological radar echo inversion data, and beam width characterization value W is the volume between adjacent range gates. Beam width characterization value W = l × h × w, where l is the length corresponding to the range gate; h is the height of the range gate, that is, the Rth n The pitch length corresponding to a range gate; w is the width of the range gate.

[0046] Construct a label matrix: determine the thresholds for different danger levels of high-altitude ice crystals, quantify the ice crystal water content (IWC) value in the airborne meteorological radar echo inversion data according to the specified threshold to obtain the high-altitude ice crystal sample label, and establish a data set D, which includes the reflectivity factor Z, temperature T, beam width representation W and the corresponding high-altitude ice crystal sample label Y.

[0047] The steps to determine the thresholds for different levels of danger of high-altitude ice crystals include: 3 For low concentration, Ag / m 3 ≤IWC 3 Medium concentration, IWC ≥ Bg / m 3 is a high concentration, where IWC is the ice crystal water content, where A represents the first threshold, B represents the second threshold, and B is greater than A.

[0048] The steps to create data set D include:

[0049] The obtained ice crystal water content IWC is quantified according to a given threshold. The specific quantification method is as follows:

[0050]

[0051] In one embodiment, A is 1 and B is 3, and the ice crystal water content IWC is quantified as follows:

[0052] ​​

[0053] The number of samples in the data set is K, and the data set D is specifically expressed as:

[0054]

[0055] Where, Y = [y1 y2…y K ] T is the quantized label vector.

[0056] The dataset D is subjected to the first stratified sampling according to the sample label Y to obtain samples of various types with a balanced number.

[0057] In the process of extracting each type of sample, the second stratified sampling is performed according to the temperature T information in the feature set to obtain a data set F with a balanced number of samples in different temperature layers.

[0058] A recognition model is established, and the data set F is used as the input of the recognition model for training to obtain a high-altitude ice crystal water content estimation model.

[0059] The actual acquired high-altitude ice crystal data of the airborne platform is input into the high-altitude ice crystal moisture content estimation model for online testing, and the high-altitude ice crystal moisture content estimation model after the test is used to estimate the high-altitude ice crystal moisture content of the airborne platform.

[0060] The method for obtaining the echo inversion data of the airborne meteorological radar includes: the echo inversion data corresponding to the kth range gate of the radar beam at the current carrier position is the average value of the meteorological data covered in the k-1th to kth range gates, and based on this logic, the azimuth and elevation scans and the traversal of different carrier positions are completed to obtain the echo inversion data of the airborne meteorological radar.

[0061] In a specific embodiment, a method for estimating the water content of high-altitude ice crystals based on an airborne platform includes:

[0062] Step 1: Obtain high-altitude ice crystal dataset:

[0063] like Figure 1 As shown, a high-altitude ice crystal data set is obtained. The data set can be obtained through numerical models or other methods, or through airborne in-situ sensors. The data is usually stored as a three-dimensional array with a known resolution. According to public information, a high-altitude ice crystal event occurred in the Cayenne region of French Guiana on May 26, 2015 UTC time. The numerical model is used to simulate this event and obtain high-altitude ice crystal grid data. The data set includes information such as radar reflectivity factor Z, temperature T, water content IWC, and three-way wind speed.

[0064] Step 2: Set the airborne weather radar echo inversion parameters:

[0065] The position of the lower left corner of the three-dimensional meteorological target is set to (0,0,0). In order to ensure that the aircraft can detect the meteorological target on the flight track, the initial position of the aircraft is set to (-1000,50000,8000). The aircraft flies at a speed of 400 knots at the initial position and at a specified heading angle. The position is updated once a minute, and the echo data update rate is also 1 time / min. The entire flight process lasts for 10 minutes. Assuming that the effective detection range of high-altitude ice crystals of the airborne meteorological radar is 80km, the azimuth and elevation beam widths of the airborne meteorological radar are both set to 4°, and the azimuth scan of ±60° and the elevation scan of ±5° are performed during the scanning process. Considering the richness and diversity of the actual detection scenes of the airborne meteorological radar, multiple flight trajectories can be planned in the simulation test, such as letting the aircraft fly at multiple heading angles such as ±45°, ±30°, and 0°.

[0066] Step 3: Obtain airborne weather radar echo inversion data

[0067] like Figure 2 As shown in the figure, assuming that the current azimuth angle is α and the pitch angle is β, the coordinates of the grid points covered by the radar beam in the k-1th to kth range gates under the current aircraft position are calculated, and the echo information of the corresponding area is extracted from the high-altitude ice crystal data set, such as the ice crystal water content [IWC1, IWC2, …, IWC M ], radar reflectivity factor [Z1,Z2,…,Z M ], etc., where M represents the number of grid points of the kth range gate of the beam covering the meteorological target. The statistical average value of each echo feature corresponding to the coverage area is used as the echo signal inverted under the kth range gate. For example, the water content of ice crystals under the kth range gate is Radar reflectivity factor In the embodiment, when other parameters are the same, the heading angles of the aircraft are set to -45°, 0° and 45° respectively. Based on the above process, the traversal of the azimuth and elevation scans under the three aircraft positions is completed to obtain the echo inversion data of the airborne meteorological radar. Note the mutual conversion between the meteorological target coordinate system, the body coordinate system and the beam coordinate system involved in the calculation process.

[0068] Step 4: Construct a feature matrix for estimating the water content of high-altitude ice crystals for airborne platforms

[0069] Taking into account the basic principles of radar detection of high-altitude ice crystals, the formation mechanism of ice crystals, and the increase in the beam coverage of radar during the detection process, the minimum feature set for identifying the water content of high-altitude ice crystals by airborne meteorological radar is constructed, namely {radar reflectivity factor Z, temperature T, beam width characterization value W}. It should be noted that in practical applications, adding other features to the minimum feature set should also fall within the scope of this right. Among them, Z and T can be obtained from the echo inversion data obtained in step three; the beam width characterization value W is a physical quantity that measures the change in beam width, which helps the identification model to establish the connection between the water content of high-altitude ice crystals and the beam width, and improve the model's ability to estimate the water content. For a conventional directional antenna, W can be expressed by two adjacent range gates R n-1 ,R n The volume between is represented by a rectangular parallelepiped, which is specifically defined as: W = l × h × w; where l is the length of the range gate; h is the height of the range gate, that is, the Rth n The pitch length corresponding to the Rth range gate; w is the width of the range gate, i.e. n Therefore, the minimum feature set for estimating the water content of high-altitude ice crystals can be recorded as:

[0070]

[0071] Where K is the total number of samples in the data set obtained in step 3. Each row in the formula corresponds to the feature vector of a sample. For example, the i-th sample x i The corresponding eigenvector is X i =[Z i T i W i ].

[0072] Step 5: Construct label matrix

[0073] According to RTCA's definition of high-altitude ice crystal hazard level, when IWC < 1g / m 3 Low concentration, 1g / m 3 ≤IWC<3g / m 3 Medium concentration, IWC ≥ 3g / m 3 Considering the difference between radar resolution and ice crystal size and the practical use of dangerous weather indications from weather radar, the ice crystal water content IWC obtained in step 3 is quantified according to the threshold value given in the above literature. The specific quantification method is as follows:

[0074]

[0075] At this point, the echo data obtained in step 3 is passed through steps 4 and 5 to obtain a new data set D. Assuming that the number of samples in the data set is K, D can be specifically expressed as:

[0076]

[0077] Where, Y = [y1 y2…y K ] T is the quantized label vector.

[0078] Step 6: Double Data Feature Sampling Method

[0079] like Figure 3 As shown in the figure, according to the statistical results of the document "Feasibility Study Weather Radar for Ice Crystal Detection", there is a significant imbalance in the number of ice crystal samples of different concentrations. In order to reduce the impact of this problem on the performance of the classification model, the data is sampled in the first stratification according to the sample label Y; even for the same type of ice crystal samples, the corresponding radar echo data volume is very large. In order to avoid the random extraction bias that still exists in the first stratification sampling, the second stratification sampling is actually performed according to the temperature T information in the feature set during the extraction of each type of sample to ensure that each type of sample can cover the feature set containing multiple temperatures. Based on the above process, the quality control of the data set D is completed, and the new data set F is obtained after double stratification sampling.

[0080] Step 7: Moisture content estimation model training

[0081] like Figure 4 As shown, a recognition model is constructed, which can be any classification algorithm such as support vector machine, random forest or neural network. The data set obtained in step six is ​​used as the input of the classification algorithm, and the hyperparameters of the algorithm are set. The optimization function is taken as the goal, and the parameter set of the classification model or the optimization function is obtained through training to obtain the high-altitude ice crystal moisture content model. Specifically, a 3-classification support vector machine SVM is constructed, each classifier uses a radial basis kernel function, and the hyperparameters of the kernel function are determined by the grid search method. The hyperparameter γ used in the embodiment is 1, and the penalty factors of 1 to 3 types of SVM are set to [0.5, 0.8, 0.9] respectively. The maximization interval between categories is taken as the optimization goal, and the parameter set of the classification model or the optimization function is obtained through training to obtain the high-altitude ice crystal moisture content model.

[0082] Step 8: Test the water content of ice crystals at high altitude on the airborne platform

[0083] During the testing phase, the acquired airborne platform high-altitude ice crystal data is input into the classification model SVM obtained in step seven for online testing to complete the prediction of the water content of the airborne platform's high-altitude ice crystals, providing necessary data input for subsequent airborne meteorological radar warnings.

[0084] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for estimating the water content of high-altitude ice crystals based on an airborne platform, characterized in that: The steps include: Obtain high-altitude ice crystal data sets as meteorological targets and set the three-dimensional position information of the meteorological targets; Establish an airborne weather radar echo inversion model, simulate the real detection scenario of the airborne weather radar, set multiple aircraft initial positions and flight paths according to the location of the weather target, set the aircraft avionics information, and set the airborne weather radar parameter information; Extract the characteristics of high-altitude ice crystals covered by the airborne weather radar beam during the flight process and obtain the airborne weather radar echo inversion data; A feature matrix for estimating the water content of high-altitude ice crystals for airborne platforms is established. The minimum feature set of the matrix is ​​the radar reflectivity factor Z, temperature T, and beam width characterization quantity W. The radar reflectivity factor Z and temperature T are obtained from the airborne meteorological radar echo inversion data, and the beam width characterization quantity W is the volume between adjacent range gates. Construct a label matrix: determine the thresholds of different danger levels of high-altitude ice crystals, quantify the ice crystal water content IWC value in the airborne meteorological radar echo inversion data according to the specified threshold to obtain the high-altitude ice crystal sample label, and establish a data set D, which includes the reflectivity factor Z, temperature T, beam width representation W and the corresponding high-altitude ice crystal sample label Y; Perform the first stratified sampling on the dataset D according to the sample label Y to obtain samples of various types with a balanced number; In the process of extracting each type of sample, the second stratified sampling is performed according to the temperature T information in the feature set to obtain a data set F with a balanced number of samples in different temperature layers; Establish a recognition model, use the data set F as the input of the recognition model for training, and obtain a high-altitude ice crystal water content estimation model; The actual acquired high-altitude ice crystal data of the airborne platform is input into the high-altitude ice crystal moisture content estimation model for online testing, and the high-altitude ice crystal moisture content estimation model after the test is used to estimate the high-altitude ice crystal moisture content of the airborne platform.

2. The method for estimating the water content of high-altitude ice crystals based on an airborne platform according to claim 1, characterized in that: The method for obtaining the echo inversion data of the airborne meteorological radar includes: the echo inversion data corresponding to the kth range gate of the radar beam at the current carrier position is the average value of the meteorological data covered in the k-1th to kth range gates, completing the azimuth and elevation scanning and the traversal of different carrier positions to obtain the echo inversion data of the airborne meteorological radar.

3. The method for estimating water content of high-altitude ice crystals based on an airborne platform according to claim 1, characterized in that: Beam width characterization quantity W = l × h × w, where l is the length corresponding to the range gate; h is the height of the range gate, that is, the Rth n The pitch length corresponding to a range gate; w is the width of the range gate.

4. The method for estimating water content of high-altitude ice crystals based on an airborne platform according to claim 1, characterized in that: Setting the aircraft avionics information includes setting the aircraft's flight speed, flight altitude, and heading angle.

5. The method for estimating water content of high-altitude ice crystals based on an airborne platform according to claim 1, characterized in that: Setting the airborne weather radar parameter information includes setting the elevation and azimuth beam widths, the elevation and azimuth scanning ranges, the maximum detection range, and the actual distance corresponding to a range gate.

6. The method for estimating the water content of high-altitude ice crystals based on an airborne platform according to claim 1, characterized in that: The steps to determine the thresholds for different levels of danger of high-altitude ice crystals include: <Ag / m 3 For low concentration, Ag / m 3 ≤IWC 3 Medium concentration, IWC ≥ Bg / m 3 is a high concentration, where IWC is the ice crystal water content, where A represents the first threshold, B represents the second threshold, and B is greater than A.​ 7. The method for estimating water content of high-altitude ice crystals based on an airborne platform according to claim 6, characterized in that: The steps to create data set D include: The obtained ice crystal water content IWC is quantified according to a given threshold. The specific quantification method is as follows: The number of samples in the data set is K, and the data set D is specifically expressed as: Where, Y = [y1 y2…y K ] T is the quantized label vector.

Images