Method, device and equipment for meteorological radar wind field inversion and storage medium

Abstract

The application discloses a meteorological radar wind field inversion method, device, equipment and storage medium, and relates to the technical field of meteorological radars. The method comprises the following steps: acquiring observation data to be processed; the observation data comprises a plurality of first direction angles on a specified distance and a first radial velocity corresponding to each first direction angle; determining whether the plurality of first radial velocities are free from velocity ambiguity by using a preset velocity ambiguity discrimination rule; in response to the plurality of first radial velocities being free from velocity ambiguity, extracting a preset number of second direction angles and a second radial velocity corresponding to each direction angle from the observation data; and obtaining a wind field inversion result by using a regularization algorithm and a wind field observation relationship model based on the plurality of second direction angles and the second radial velocity.

Description

Technical Field

[0001] This application relates to the field of meteorological radar technology, specifically to wind field detection technology, meteorological detection technology and other technical fields, and particularly to a method, apparatus, equipment and storage medium for wind field inversion by meteorological radar. Background Technology

[0002] With the development of meteorological radar technology, using radar radial velocity observation to invert horizontal two-dimensional wind fields has become an important means of mesoscale weather analysis and severe convection monitoring.

[0003] Currently, relevant technical solutions include wind field inversion methods based on azimuth scanning volumetric velocity processing techniques, such as wind field inversion methods based on the Velocity Azimuth Display (VAD) algorithm. However, the VAD algorithm has high requirements for the distribution density of the observed azimuth angle and relies on relatively complete or high-density radar scan data, which faces significant limitations in practical applications. Especially when radar observation is limited, scanning is incomplete, obstruction is severe, or some azimuth observations are missing, it is often difficult to obtain stable and reliable wind field estimation results. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and storage medium for wind field inversion using meteorological radar, which can solve the problem of poor accuracy in wind field estimation results under complex observation scenarios. The technical solution is as follows:

[0005] Firstly, a method for retrieving wind fields from a weather radar is provided, the method comprising:

[0006] Acquire the observation data to be processed; the observation data includes multiple first direction angles at a specified distance and a first radial velocity corresponding to each first direction angle;

[0007] Using a preset velocity ambiguity discrimination rule, determine whether multiple first radial velocities are free from velocity ambiguity;

[0008] In response to the absence of velocity ambiguity among multiple first radial velocities, a predetermined number of second orientation angles and the second radial velocity corresponding to each orientation angle are extracted from the observation data;

[0009] Based on multiple second direction angles and second radial velocities, wind field inversion results are obtained using a regularization algorithm and a wind field observation relationship model.

[0010] In one possible implementation, determining whether multiple first radial velocities are free from velocity ambiguity using a preset velocity ambiguity discrimination rule includes:

[0011] Based on multiple first radial velocities, determine the mean absolute error and standard deviation corresponding to the multiple first radial velocities;

[0012] Based on the mean absolute error and standard deviation, and using a preset velocity fuzziness discrimination rule, it is determined whether multiple first radial velocities are free from velocity fuzziness.

[0013] In one possible implementation, determining whether multiple first radial velocities are free from velocity ambiguity based on the mean absolute error and standard deviation using a preset velocity ambiguity discrimination rule includes:

[0014] Based on the preset boundary function and the mean absolute error, the standard deviation boundary threshold is calculated.

[0015] In response to the standard deviation being greater than the standard deviation boundary threshold, it is determined that there is no velocity ambiguity among the multiple first radial velocities;

[0016] In response to the standard deviation being less than or equal to the standard deviation boundary threshold, velocity ambiguity is determined to exist in a plurality of first radial velocities.

[0017] In one possible implementation, the wind field inversion result is obtained based on multiple second direction angles and second radial velocities, using a regularization algorithm and a wind field observation relationship model, including:

[0018] Based on the preset L-curve criterion, the regularization parameters of the regularization algorithm to be optimized are optimized to determine the regularization algorithm.

[0019] Based on multiple second direction angles and second radial velocities, the wind field inversion results are obtained using the regularization algorithm and the wind field observation relationship model.

[0020] In one possible implementation, obtaining the wind field inversion result based on the wind field observation relationship model using the regularization algorithm, based on multiple second direction angles and second radial velocities, includes:

[0021] Based on multiple second direction angles and second radial velocities, the wind speed components are calculated using the wind field observation relationship model.

[0022] The wind speed component is adjusted to a preset range using the regularization algorithm to obtain the target wind speed component;

[0023] The wind field inversion result is obtained based on the target wind speed component.

[0024] In one possible implementation, the optimization of the regularization parameters of the regularization algorithm to be optimized based on a preset L-curve criterion, to determine the regularization algorithm, includes:

[0025] Based on the objective function of the regularization algorithm to be optimized, the relationship curve between the data residual norm and the regularization term norm in the objective function is determined;

[0026] Based on the relationship curve between the residual norm and the regularization term norm of the data, the inflection point of the curve is determined;

[0027] The regularization parameter corresponding to the inflection point of the curve is used as the regularization parameter after optimization.

[0028] Based on the optimized regularization parameters and objective function, the optimized regularization algorithm is obtained.

[0029] Secondly, a device for retrieving wind fields from a weather radar is provided, the device comprising:

[0030] An acquisition unit is used to acquire observation data to be processed; the observation data includes multiple first direction angles at a specified distance and a first radial velocity corresponding to each first direction angle.

[0031] The determining unit is used to determine whether multiple first radial velocities are free from velocity ambiguity using a preset velocity ambiguity discrimination rule;

[0032] An extraction unit is configured to extract a preset number of second direction angles and the second radial velocity corresponding to each direction angle from the observation data in response to the absence of velocity ambiguity in multiple first radial velocities.

[0033] The unit is used to obtain wind field inversion results based on multiple second direction angles and second radial velocities, using a regularization algorithm and a wind field observation relationship model.

[0034] Thirdly, a computer-readable storage medium is provided, wherein at least one instruction is stored therein, the at least one instruction being loaded and executed by a processor to implement the aspects and any possible implementations described above.

[0035] Fourthly, an electronic device is provided, comprising:

[0036] At least one processor; and

[0037] A memory communicatively connected to the at least one processor; wherein,

[0038] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods described above and any possible implementations.

[0039] The beneficial effects of the technical solution provided in this application include at least the following:

[0040] As can be seen from the above technical solution, the embodiments of this application can obtain observation data to be processed; the observation data includes multiple first azimuth angles at a specified distance and a first radial velocity corresponding to each first azimuth angle. Using a preset velocity ambiguity discrimination rule, it is determined whether the multiple first radial velocities are free from velocity ambiguity. In response to the absence of velocity ambiguity in the multiple first radial velocities, a preset number of second azimuth angles and a second radial velocity corresponding to each azimuth angle are extracted from the observation data. Based on the multiple second azimuth angles and second radial velocities, a regularization algorithm and a wind field observation relationship model are used to obtain the wind field inversion result. Since the wind field inversion result can be directly obtained by using the radial velocity and azimuth angle of the observation data without velocity ambiguity, the dependence on complete and dense radar observation data can be avoided. The wind field inversion can be efficiently completed based on only a limited number of radar observation data. This can effectively address the practical problems such as incomplete observation data or limited scanning, improve the practicality of the algorithm under complex meteorological conditions and limited observation resources, effectively improve the performance of wind field parameter inversion processing in complex meteorological environments, and thus ensure the reliability of the wind field inversion result of meteorological radar.

[0041] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic flowchart of a method for retrieving wind fields from a weather radar according to an embodiment of this application;

[0044] Figure 2 This is a schematic diagram of the scatter plot and boundary function in the weather radar wind field inversion method provided in one embodiment of this application;

[0045] Figure 3 This is a schematic diagram of the result obtained based on the L-curve criterion in the method for inverting wind fields using weather radar provided in one embodiment of this application;

[0046] Figure 4 This is a structural block diagram of a weather radar wind field inversion device provided in another embodiment of this application;

[0047] Figure 5This is a block diagram of an electronic device used to implement the weather radar wind field inversion method of the embodiments of this application. Detailed Implementation

[0048] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0049] Obviously, the described embodiments are only some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0050] It should be noted that the hardware observation equipment terminal involved in the embodiments of this application may include, but is not limited to, equipment in meteorological observation stations such as receivers, integrated processors, remote control devices, and base measurement boxes.

[0051] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0052] With the development of meteorological radar technology, using radar radial velocity observation to invert horizontal two-dimensional wind fields has become an important means of mesoscale weather analysis and severe convection monitoring.

[0053] Currently, relevant technical solutions include wind field inversion methods based on azimuth scanning volumetric velocity processing techniques, such as wind field inversion methods based on the VAD algorithm. This approach is simple to implement and widely used, dominating radar wind field inversion. However, the VAD algorithm has high requirements for the distribution density of the observation azimuth angle, relying on relatively complete or high-density radar scan data, which faces significant limitations in practical applications. Especially when radar observation is limited, scanning is incomplete, obstruction is severe, or some azimuth observations are missing, it is often difficult to obtain stable and reliable wind field estimation results.

[0054] Furthermore, radar observation data is often accompanied by strong noise and anomalous disturbances due to weather conditions, electromagnetic interference, and systematic errors, further exacerbating the ill-conditioned nature and uncertainty of the wind field inversion problem. Especially when there is limited or low-quality observation data, the inversion process based on the least squares method is prone to producing unstable solutions or physically unreasonable estimation results, seriously affecting the practicality and reliability of the wind field inversion algorithm.

[0055] To address the aforementioned issues, there is an urgent need for a wind field inversion method that maintains good stability, robustness, and wide adaptability under low observation conditions, breaking through the dependence of traditional algorithms on the completeness of observations and enhancing the application capability of radar wind field inversion technology in complex meteorological environments.

[0056] Therefore, there is an urgent need for a method for wind field inversion from meteorological radar to effectively invert wind field parameters in complex meteorological environments, thereby ensuring the reliability of the wind field inversion results.

[0057] Please refer to Figure 1 This document illustrates a flowchart of a method for retrieving wind fields from a weather radar according to an embodiment of this application. Specifically, this method may include:

[0058] Step 101: Obtain the observation data to be processed; the observation data includes multiple first direction angles at a specified distance and a first radial velocity corresponding to each first direction angle.

[0059] Step 102: Using the preset velocity fuzziness discrimination rules, determine whether multiple first radial velocities are free from velocity fuzziness.

[0060] Step 103: In response to the absence of velocity ambiguity in the multiple first radial velocities, extract a preset number of second direction angles and the second radial velocity corresponding to each direction angle from the observation data.

[0061] Step 104: Based on multiple second direction angles and second radial velocities, wind field inversion results are obtained using a regularization algorithm and a wind field observation relationship model.

[0062] It should be noted that weather radar can include S-band radar, etc. Wind fields can include two-dimensional horizontal wind fields.

[0063] It should be noted that the observation data may include multiple distance libraries at a specified distance. Each distance library includes the orientation angle and radial velocity.

[0064] It should be noted that the specified distance can be the radius length centered on the weather radar. There can be multiple specified distances. For example, it could be multiple specified distances within a radius range of 10 to 50 kilometers at the third elevation angle of the weather radar.

[0065] It should be noted that the wind field inversion results may include horizontal wind speed and wind direction angle.

[0066] It should be noted that the preset number can be a single-digit number. For example, the preset number can be 8.

[0067] In this way, the observation data to be processed can be acquired. The observation data includes multiple first azimuth angles at a specified distance and a first radial velocity corresponding to each first azimuth angle. Using a preset velocity ambiguity discrimination rule, it is determined whether the multiple first radial velocities are free from velocity ambiguity. In response to the absence of velocity ambiguity in the multiple first radial velocities, a preset number of second azimuth angles and a second radial velocity corresponding to each azimuth angle are extracted from the observation data. Based on the multiple second azimuth angles and second radial velocities, a regularization algorithm and a wind field observation relationship model are used to obtain the wind field inversion result. Since the wind field inversion result can be directly obtained by using the radial velocity and azimuth angle of the observation data without velocity ambiguity, the dependence on complete and dense radar observation data can be avoided. The wind field inversion can be efficiently completed based on only a limited number of radar observation data. This can effectively address the practical problems such as incomplete observation data or limited scanning, improve the practicality of the algorithm under complex meteorological conditions and limited observation resources, and effectively improve the performance of wind field parameter inversion processing in complex meteorological environments, thereby ensuring the reliability of the wind field inversion result of meteorological radar.

[0068] Optionally, in one possible implementation of this embodiment, in step 102, firstly, the mean absolute error and standard deviation corresponding to the multiple first radial velocities can be determined based on the multiple first radial velocities. Secondly, based on the mean absolute error and standard deviation, a preset velocity fuzziness discrimination rule can be used to determine whether the multiple first radial velocities are free from velocity fuzziness.

[0069] In this implementation, the standard deviation of the multiple first radial velocities can be the standard deviation of the absolute errors of the multiple first radial velocities.

[0070] In a specific implementation of this method, firstly, a preset fitting algorithm can be used to perform curve fitting on multiple first radial velocities to obtain a fitted curve for the first radial velocity. Secondly, based on the fitted curve and the multiple first radial velocities, the absolute error corresponding to the multiple first radial velocities can be calculated. Thirdly, the mean absolute error and standard deviation of the absolute errors are calculated. Finally, the mean absolute error and standard deviation of the absolute errors are used as the mean absolute error and standard deviation corresponding to the multiple first radial velocities.

[0071] Here, the preset fitting algorithm may include the VAD algorithm, etc.

[0072] In another specific implementation of this approach, firstly, a standard deviation boundary threshold can be calculated based on a preset boundary function and the mean absolute error. In response to the standard deviation being greater than the standard deviation boundary threshold, it is determined that the multiple first radial velocities do not exhibit velocity ambiguity; or, in response to the standard deviation being less than or equal to the standard deviation boundary threshold, it is determined that the multiple first radial velocities exhibit velocity ambiguity.

[0073] Preferably, the preset boundary function can be as shown in formula (1):

[0074] (1)

[0075] in, The standard deviation threshold, This represents the mean absolute error.

[0076] Here, firstly, the average absolute error corresponding to multiple first radial velocities can be substituted into formula (1) to calculate the corresponding standard deviation boundary threshold. Secondly, the standard deviations corresponding to multiple first radial velocities and the standard deviation boundary threshold can be compared to determine whether the standard deviations corresponding to multiple first radial velocities are greater than the standard deviation boundary threshold. If the standard deviation is greater than the standard deviation boundary threshold, it can be determined that there is no velocity ambiguity among the multiple first radial velocities at the specified distance. If the standard deviation is less than or equal to the standard deviation boundary threshold, it can be determined that there is velocity ambiguity among the multiple first radial velocities at the specified distance.

[0077] In this implementation, the preset boundary function can be pre-constructed based on velocity fuzzy sample data and non-velocity fuzzy sample data.

[0078] In another specific implementation of this method, before step 102, firstly, based on the velocity-fuzzy sample data and the non-velocity-fuzzy sample data, curve fitting is performed using the VAD algorithm to obtain velocity-fuzzy fitting curves and non-velocity-fuzzy fitting curves respectively. Secondly, the velocity-fuzzy absolute error is calculated based on the velocity-fuzzy fitting curves and the corresponding original radial velocities, and the non-velocity-fuzzy absolute error is calculated based on the non-velocity-fuzzy fitting curves and the original radial velocities in the corresponding sample data. Thirdly, the velocity-fuzzy mean absolute error and velocity-fuzzy standard deviation of the velocity-fuzzy absolute error are calculated. Fourthly, the non-velocity-fuzzy mean absolute error and non-velocity-fuzzy standard deviation of the non-velocity-fuzzy absolute error are calculated. Finally, a scatter plot is plotted based on the velocity-fuzzy mean absolute error and velocity-fuzzy standard deviation, and the non-velocity-fuzzy mean absolute error and non-velocity-fuzzy standard deviation. Based on the scatter plot, linear fitting processing is performed to obtain a preset boundary function.

[0079] In the specific implementation, velocity-fuzzy sample data can include an orientation angle and the corresponding radial velocity. Non-velocity-fuzzy sample data can also include an orientation angle and the corresponding radial velocity. Here, the following operations can be performed on both velocity-fuzzy and non-velocity-fuzzy sample data: First, based on the orientation angle and the corresponding radial velocity in the sample data, a fitting curve can be obtained using the VAD algorithm, and the fitted radial velocity can be calculated. Second, the absolute error between the fitted radial velocity and the radial velocity in the sample data is calculated. Third, based on the absolute error, the mean absolute error and standard deviation of the absolute error are calculated.

[0080] In one specific implementation case, the VAD algorithm can be as shown in formula (2):

[0081] (2)

[0082] in, Let v be the east-west wind velocity vector, and v be the north-south wind velocity vector. Let i be the azimuth angle. The radial velocity at the i-th azimuth angle at a specified distance r.

[0083] For example, to determine velocity ambiguity, the Mean Absolute Error (MAE) and Standard Deviation (SD) can be used as velocity ambiguity discrimination indicators to preset boundary functions. First, multiple sets of sample data without velocity ambiguity can be obtained, for example, 1920 sets of sample data without velocity ambiguity, and multiple sets of sample data with velocity ambiguity can be obtained, for example, 2322 sets of sample data with velocity ambiguity, for statistical analysis. Second, using the VAD algorithm, curve fitting is performed on each set of sample data to obtain a fitted curve. On one hand, the absolute error between the fitted curve and the total original radial velocity of the sample data at each azimuth angle is calculated, and the average of these errors is used to obtain the MAE. On the other hand, the SD of the absolute error is calculated. Based on this, the discrimination indicators between ambiguous and unambiguous samples are statistically analyzed, and a scatter plot is plotted. Figure 2 This is a scatter plot and a schematic diagram of the boundary function in a weather radar wind field inversion method provided in one embodiment of this application, as shown below. Figure 2As shown, unambiguous data and fuzzy data form a boundary, and the boundary function shown by the dashed line can be fitted. The upper part of the dashed line is unambiguous data, and the lower part is fuzzy data. Unambiguous data refers to data points where there is no velocity fuzziness sample data corresponding to the MAE and SD. Fuzzy data refers to data points where there is velocity fuzziness sample data corresponding to the MAE and SD. There is a clear boundary between the two types of sample data in the feature space of MAE and SD. Further, the boundary line function is determined by linear fitting, that is, the preset boundary function, as shown in formula (1), thereby establishing the velocity fuzziness discrimination rule. The preset boundary function formula can be used to quickly determine whether there is velocity fuzziness in the radial velocity in the observation data, and assist in screening radial velocity data that meets the requirements, providing reliable support for subsequent inversion.

[0084] Here, a set of sample data may include multiple orientation angles and the corresponding radial velocity for each orientation angle. A set of sample data may include multiple pairs of orientation angle and radial velocity data.

[0085] Understandably, radial velocity information at a fixed elevation angle and corresponding to different azimuth angles at a given range is extracted from radar echo data. Here, for the radial velocity of each range database, it should be ensured that there is no velocity ambiguity, or that any ambiguity has been resolved to ensure that the data can be used for inversion analysis.

[0086] In this way, by using the mean absolute error and standard deviation corresponding to multiple first radial velocities and employing a preset velocity fuzziness discrimination rule, it is possible to accurately and effectively determine whether there is velocity fuzziness at the specified distance for the first radial velocity. This allows for subsequent wind field parameter inversion processing based on the first radial velocity without velocity fuzziness, further ensuring the reliability of the wind field inversion results.

[0087] Optionally, in one possible implementation of this embodiment, in step 104, firstly, based on a preset L-curve criterion, the regularization parameters of the regularization algorithm to be optimized are optimized to determine the regularization algorithm. Secondly, based on multiple second direction angles and second radial velocities, the wind field inversion results can be obtained using the regularization algorithm and the wind field observation relationship model.

[0088] In this implementation, the regularization algorithm may include Tikhonov regularization, Lasso regularization, Elastic Net regularization, Total Variation regularization, and other algorithms.

[0089] In a specific implementation of this method, firstly, wind speed components can be calculated using the wind field observation relationship model based on multiple second direction angles and second radial velocities. Secondly, the wind speed components are adjusted to a preset range using the regularization algorithm to obtain the target wind speed component. Thirdly, the wind field inversion result is obtained based on the target wind speed component.

[0090] In this implementation, the wind field observation relationship model can be a pre-constructed set of observation equations. The regularization algorithm can be an optimized version of the regularization algorithm.

[0091] For example, at a specified distance r, there are typically more than 360 distance databases, usually sampled at 1° or finer angular intervals. Multiple second azimuth angles and second radial velocities can be the radial velocities of the distance databases at eight different azimuth angles. For the selected radial velocities, an observation model for wind field inversion, i.e., a wind field observation relationship model, can be pre-established.

[0092] Here, we can set the two-dimensional wind speed components to be inverted as u and v, and for each azimuth angle Corresponding radial velocity By organizing the observation relationships of all n azimuth angles into a matrix form, the observation equation set can be constructed as shown in formula (3):

[0093] (3)

[0094] in, Let A be an n×1 observation vector; let A be an n×2 observation matrix, with the i-th row being... , Let be the two-dimensional wind speed component vector to be determined.

[0095] Because the observation equations often exhibit ill-conditioned matrix characteristics due to observation noise, direct solutions frequently lead to unstable results and incorrect numerical values. To overcome this problem, a regularization algorithm can be used to restrict the solutions of the wind speed components to a certain norm range. The objective function of the regularization algorithm is shown in formula (4):

[0096] (4)

[0097] in, L is the regularization parameter; L is usually the identity matrix, representing a constraint on the overall magnitude of the solution.

[0098] Based on multiple second direction angles and second radial velocities, the final wind speed components can be solved using formulas (3) and (4). , Then, the final horizontal wind speed can be calculated using a preset speed algorithm, and the wind direction angle can be calculated using a preset wind direction angle algorithm. The horizontal wind speed and wind direction angle can be used as the two-dimensional horizontal wind field inversion result at that distance.

[0099] Here, the preset speed algorithm can be as shown in formula (5):

[0100] (5)

[0101] in, For horizontal wind speed, The wind velocity vector runs in the east-west direction. It can be the wind velocity vector in the north-south direction.

[0102] The preset wind direction angle algorithm can be shown in formula (6):

[0103] (6)

[0104] in, Wind direction angle It can be the east-west wind velocity vector. It can be the wind velocity vector in the north-south direction.

[0105] In another specific implementation of this approach, firstly, based on the objective function of the regularization algorithm to be optimized, the relationship curve between the data residual norm and the regularization term norm in the objective function can be determined. Secondly, based on the relationship curve between the data residual norm and the regularization term norm, the inflection point of the curve can be determined. Thirdly, the regularization parameter corresponding to the inflection point of the curve is used as the optimized regularization parameter. Finally, based on the optimized regularization parameter and the objective function, the optimized regularization algorithm is obtained.

[0106] One implementation method involves obtaining multiple preset regularization parameters. Based on these parameters and the objective function of the regularization algorithm to be optimized, a solution to the objective function can be calculated. Then, based on the solution, a curve showing the relationship between the norm of the data residuals and the norm of the regularization term in the objective function on logarithmic coordinates is plotted. Here, the points on this curve can represent regularization parameters.

[0107] Another approach to this implementation involves: First, determining the beginning and end points of the curve relating the data residual norm and the regularization term norm. Second, calculating the angles formed by the beginning and end points of the curve with every other point on the curve, resulting in multiple angles. Third, determining the minimum angle among these angles. Finally, using the regularization parameter of the point on the curve corresponding to the minimum angle as the optimized regularization parameter.

[0108] For example, the regularization algorithm can be the Tikhonov regularization algorithm. Under the Tikhonov regularization framework, its objective function can be as shown in formula (4), and the solution... By minimizing the data fitting residuals and regular term residuals The objective function is obtained by combining the components.

[0109] It is understandable that for each given regularization parameter... Each of them corresponds to a unique solution. This solution guarantees the accuracy of the observation matrix. While fitting the solution, it is also constrained by the regularization term, resulting in a certain degree of smoothness in the solution. With... Changes, solutions A balance will be struck between stability and data fit: smaller It tends to fit the original data more accurately, but it easily amplifies noise, leading to unstable solutions; larger... It can significantly suppress oscillations and noise in the solution, but may lose useful features in the data. Therefore, choosing a suitable regularization parameter is crucial. Here, the L-curve criterion can be used to obtain the optimal regularization parameters. We can first calculate a series of different Corresponding solution Then, plot the relationship between the data residual norm and the regularization term norm on logarithmic coordinates. The curve usually exhibits a typical "L"-shaped structure, where the horizontal segment corresponds to the small... The underregularized state, the vertical segment corresponds to a large The over-regularized state is represented by the curve's corner, which indicates the optimal trade-off between residuals and regularization. Ultimately, by selecting the corner... This allows us to obtain a regularized solution that is both stable and can fit the data well. .

[0110] Figure 3 This is a schematic diagram of the result obtained based on the L-curve criterion in the meteorological radar wind field inversion method provided in another embodiment of this application, as shown below. Figure 3 As shown, the blue curve represents the relationship between the data residual norm and the regularization term norm, determined based on the L-curve criterion. The points on the curve can represent the regularization parameters. Preferably, the regularization parameter corresponding to the inflection point of the curve can be determined based on the L-curve criterion and the relationship curve between the data residual norm and the regularization term norm. Given the positions of the first and last points of the curve, calculate the angles formed by the first and last points and each regularization parameter, and find the minimum angle. The corresponding point is the inflection point of the curve. This can be the optimal regularization parameter. Selected. Then, substitute it into the objective function formula (4) to determine the objective function of the optimized regularization algorithm, and then determine the optimized regularization algorithm to solve for the final wind speed component. , .

[0111] In this way, the horizontal two-dimensional wind field can be inverted by selecting a small amount of radar radial velocity observation data, which greatly reduces the dependence on the amount of observation data and effectively overcomes the limitation of related technical solutions that require complete or high-density azimuth observations, thus improving the applicability of the method in scenarios with missing data or incomplete scanning.

[0112] Furthermore, the scheme in this embodiment can significantly improve the stability and robustness of the inversion. By introducing the Tikhonov regularization method, this invention can stably solve ill-conditioned or underdetermined problems in the wind field inversion process, suppress abnormal solutions caused by observation noise, and ensure the numerical stability of the inversion results.

[0113] Furthermore, the scheme in this embodiment can enhance the broad adaptability of the wind field inversion algorithm: The present invention is theoretically applicable to any wind field inversion scenario without velocity ambiguity, and has better stability and recovery capability, especially under observation obstruction, extreme weather or low observation quality conditions, thus broadening the application boundaries of radar wind field inversion technology.

[0114] It should be noted that the specific implementation process provided in this embodiment can be combined with various specific implementation processes provided in the aforementioned implementation methods to realize the meteorological radar wind field inversion method of this embodiment. Detailed descriptions can be found in the relevant content of the aforementioned implementation methods, and will not be repeated here.

[0115] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0116] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0117] Figure 4 This application provides a structural block diagram of a weather radar wind field inversion apparatus according to one embodiment of the present application. Figure 4 As shown. The weather radar wind field inversion device 400 of this embodiment may include an acquisition unit 401, a determination unit 402, an extraction unit 403, and an acquisition unit 404. The acquisition unit 401 is used to acquire observation data to be processed; the observation data includes multiple first azimuth angles at a specified distance and a first radial velocity corresponding to each first azimuth angle; the determination unit 402 is used to determine whether the multiple first radial velocities are free from velocity ambiguity using a preset velocity ambiguity discrimination rule; the extraction unit 403 is used to extract a preset number of second azimuth angles and a second radial velocity corresponding to each azimuth angle from the observation data in response to the absence of velocity ambiguity in the multiple first radial velocities; the acquisition unit 404 is used to obtain the wind field inversion result based on the multiple second azimuth angles and second radial velocities, using a regularization algorithm and a wind field observation relationship model.

[0118] Optionally, in one possible implementation of this embodiment, the determining unit 402 is used to determine the mean absolute error and standard deviation corresponding to the multiple first radial velocities based on the multiple first radial velocities; and based on the mean absolute error and standard deviation, to determine whether the multiple first radial velocities are free from velocity ambiguity using a preset velocity ambiguity discrimination rule.

[0119] Optionally, in one possible implementation of this embodiment, the determining unit 402 is used to calculate a standard deviation boundary threshold based on a preset boundary function and the mean absolute error; in response to the standard deviation being greater than the standard deviation boundary threshold, determine that there is no velocity ambiguity in the plurality of first radial velocities; in response to the standard deviation being less than or equal to the standard deviation boundary threshold, determine that there is velocity ambiguity in the plurality of first radial velocities.

[0120] Optionally, in one possible implementation of this embodiment, the obtaining unit 404 is used to optimize the regularization parameters of the regularization algorithm to be optimized based on a preset L-curve criterion, so as to determine the regularization algorithm; and to obtain the wind field inversion result based on multiple second direction angles and second radial velocities, using the regularization algorithm and the wind field observation relationship model.

[0121] Optionally, in one possible implementation of this embodiment, the obtaining unit 404 is used to calculate the wind speed component based on multiple second direction angles and second radial velocities using the wind field observation relationship model; adjust the wind speed component to a preset range using the regularization algorithm to obtain a target wind speed component; and obtain the wind field inversion result based on the target wind speed component.

[0122] Optionally, in one possible implementation of this embodiment, the obtaining unit 404 is used to determine the relationship curve between the data residual norm and the regularization term norm in the objective function of the regularization algorithm to be optimized; determine the inflection point of the curve based on the relationship curve between the data residual norm and the regularization term norm; use the regularization parameter corresponding to the inflection point of the curve as the optimized regularization parameter; and obtain the optimized regularization algorithm based on the optimized regularization parameter and the objective function.

[0123] In this embodiment, the observation data to be processed can be acquired by the acquisition unit. The observation data includes multiple first azimuth angles at a specified distance and a first radial velocity corresponding to each first azimuth angle. The determination unit uses a preset velocity ambiguity discrimination rule to determine whether the multiple first radial velocities are free from velocity ambiguity. In response to the absence of velocity ambiguity in the multiple first radial velocities, the extraction unit extracts a preset number of second azimuth angles and a second radial velocity corresponding to each azimuth angle from the observation data. The acquisition unit uses a regularization algorithm and a wind field observation relationship model based on the multiple second azimuth angles and second radial velocities to obtain the wind field inversion result. Since the wind field inversion result can be directly obtained by using the radial velocity and azimuth angle of the observation data without velocity ambiguity, the dependence on complete and dense radar observation data can be avoided. The wind field inversion can be efficiently completed based on only a limited number of radar observation data. This can effectively address the practical problems such as incomplete observation data or limited scanning, improve the practicality of the algorithm under complex meteorological conditions and limited observation resources, and effectively improve the performance of wind field parameter inversion processing in complex meteorological environments, thereby ensuring the reliability of the wind field inversion result of meteorological radar.

[0124] The processing of user personal information, such as user image and attribute data, including collection, storage, use, processing, transmission, provision, and disclosure, as well as the processing of meteorological observation data and meteorological operational parameters, involved in the technical solution of this application, all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0125] According to embodiments of this application, this application also provides an electronic device, a readable storage medium, and a computer program product.

[0126] Figure 5A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of this application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.

[0127] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0128] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, such as keyboard, mouse, etc.; output unit 507, such as various types of monitors, speakers, etc.; storage unit 508, such as disk, optical disk, etc.; and communication unit 509, such as network card, modem, wireless transceiver, etc. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0129] The computing unit 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the method for inverting weather radar wind fields. For example, in some embodiments, the method for inverting weather radar wind fields can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the method for inverting weather radar wind fields described above can be performed. Alternatively, in other embodiments, computing unit 501 may be configured by any other suitable means (e.g., by means of firmware) to perform a method for inverting weather radar wind fields.

[0130] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0131] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0132] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0133] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0134] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0135] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0136] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0137] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for wind field inversion using meteorological radar, characterized in that, The method includes: Acquire the observation data to be processed; the observation data includes multiple first direction angles at a specified distance and a first radial velocity corresponding to each first direction angle; Using a preset velocity ambiguity discrimination rule, determine whether multiple first radial velocities are free from velocity ambiguity; In response to the absence of velocity ambiguity among multiple first radial velocities, a predetermined number of second orientation angles and the second radial velocity corresponding to each orientation angle are extracted from the observation data; Based on multiple second orientation angles and second radial velocities, wind field inversion results are obtained using a regularization algorithm and a wind field observation relationship model; among which, The step of determining whether multiple first radial velocities are free from velocity ambiguity using a preset velocity ambiguity discrimination rule includes: Based on multiple first radial velocities, determine the mean absolute error and standard deviation corresponding to the multiple first radial velocities; Based on the preset boundary function and the mean absolute error, the standard deviation boundary threshold is calculated. In response to the standard deviation being greater than the standard deviation boundary threshold, it is determined that there is no velocity ambiguity among the multiple first radial velocities; In response to the standard deviation being less than or equal to the standard deviation boundary threshold, velocity ambiguity is determined to exist in a plurality of first radial velocities.

2. The method according to claim 1, characterized in that, The wind field inversion results are obtained based on multiple second direction angles and second radial velocities, using a regularization algorithm and a wind field observation relationship model, including: Based on the preset L-curve criterion, the regularization parameters of the regularization algorithm to be optimized are optimized to determine the regularization algorithm. Based on multiple second direction angles and second radial velocities, the wind field inversion results are obtained using the regularization algorithm and the wind field observation relationship model.

3. The method according to claim 2, characterized in that, The wind field inversion results are obtained based on multiple second direction angles and second radial velocities, using the regularization algorithm and wind field observation relationship model, including: Based on multiple second direction angles and second radial velocities, the wind speed components are calculated using the wind field observation relationship model. The wind speed component is adjusted to a preset range using the regularization algorithm to obtain the target wind speed component; The wind field inversion result is obtained based on the target wind speed component.

4. The method according to claim 2, characterized in that, The optimization of the regularization parameters of the regularization algorithm to be optimized, based on the preset L-curve criterion, to determine the regularization algorithm includes: Based on the objective function of the regularization algorithm to be optimized, the relationship curve between the data residual norm and the regularization term norm in the objective function is determined; Based on the relationship curve between the residual norm and the regularization term norm of the data, the inflection point of the curve is determined; The regularization parameter corresponding to the inflection point of the curve is used as the regularization parameter after optimization. Based on the optimized regularization parameters and objective function, the optimized regularization algorithm is obtained.

5. A device for wind field inversion using weather radar, characterized in that, The device includes: An acquisition unit is used to acquire observation data to be processed; the observation data includes multiple first direction angles at a specified distance and a first radial velocity corresponding to each first direction angle. The determining unit is used to determine whether multiple first radial velocities are free from velocity ambiguity using a preset velocity ambiguity discrimination rule; An extraction unit is configured to extract a preset number of second direction angles and the second radial velocity corresponding to each direction angle from the observation data in response to the absence of velocity ambiguity in multiple first radial velocities. The acquisition unit is used to obtain wind field inversion results based on multiple second direction angles and second radial velocities, utilizing a regularization algorithm and a wind field observation relationship model; among which, The determining unit is configured to determine the mean absolute error and standard deviation corresponding to the multiple first radial velocities based on the multiple first radial velocities; calculate the standard deviation boundary threshold based on the preset boundary function and the mean absolute error; determine that there is no velocity ambiguity among the multiple first radial velocities in response to the standard deviation being greater than the standard deviation boundary threshold; and determine that there is velocity ambiguity among the multiple first radial velocities in response to the standard deviation being less than or equal to the standard deviation boundary threshold.

6. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 4.

8. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-4.

Images