Single dual-polarization radar data self-filling method based on physical constraint and dynamic optimization

Abstract

The invention discloses a single dual-polarization radar data self-filling method based on physical constraint and dynamic optimization, and relates to the technical field of data filling. Comprising the following steps: acquiring original observation data, preprocessing the original observation data, and calculating a beam blocking rate; establishing a reliability criterion, and selecting an optimal filling path for each pixel point; when the relative differential phase is determined to be reliable, constructing a core path based on a-relationship; when the differential phase is not reliable but the differential reflectivity is reliable, an auxiliary path is constructed based on a statistical relationship, and data filling is executed hierarchically; and solving a reflectivity field which is finally filled and is physically coherent by minimizing a cost function. Smooth transition of the filling area and the surrounding effective observation area is ensured, and a high-quality data field is formed.

Description

Technical Field

[0001] This invention relates to the field of data filling technology, and more specifically to a self-filling method for single-unit dual-polarization radar data based on physical constraints and dynamic optimization. Background Technology

[0002] Weather radar, as a key tool for monitoring precipitation, plays an irreplaceable role in meteorology and hydrology. However, in mountainous and hilly areas with complex terrain, the radar beam propagation path is often blocked by mountains, resulting in a severe beam jamming effect. This jamming leads to data gaps or significant underestimation in the blocked areas, creating monitoring blind spots and greatly affecting the effectiveness of radar data in mountain flood warnings and geological disaster prevention. Currently, the main technical approach to solving the beam jamming problem is to use multi-radar data fusion. However, this method requires a sufficiently dense radar observation network with sufficient overlap in the coverage areas of each radar.

[0003] In practical applications, the deployment of radar networks has significant limitations. Due to factors such as construction costs, terrain constraints, and maintenance resources, many areas, especially mountainous, border, and remote regions, often only have a single radar providing monitoring coverage. In such areas with only a single dual-polarization radar site, traditional multi-radar composite technologies cannot be implemented, making the development of single-site data restoration methods that do not rely on radar networks urgently needed.

[0004] The maturity of dual-polarization radar technology offers new possibilities for solving this problem. Unlike traditional single-polarization radar, which can only acquire reflectivity factor, dual-polarization radar can simultaneously acquire multiple observation parameters, such as reflectivity factor, differential reflectivity, differential phase ratio, and correlation coefficient, by transmitting and receiving electromagnetic waves with both horizontal and vertical polarization directions. These parameters are closely physically correlated, providing a physical basis for data self-repair. Among them, the differential phase ratio, as a physical quantity characterizing the shape and concentration of particles along the beam propagation path, is insensitive to beam jamming. When the radar beam is partially blocked by terrain, the reflectivity factor is significantly underestimated, while the differential phase ratio still maintains good observation quality, making it an ideal benchmark parameter for data repair in beam-blocked areas.

[0005] In beam-blocking scenarios, compared to other observables, horizontal reflectivity ( Horizontal reflectivity is the most easily lost and faulty of all dual-polarization parameters. It measures the energy of the horizontally polarized wave returned by a target; it is an absolute measurement, appearing on radar images behind beam-blocking areas. Values ​​may be significantly underestimated, or filtered out due to weak signals, resulting in data gaps. However, the horizontal reflectivity factor is a fundamental radar observation product, serving not only as a crucial input for quantitative precipitation estimation but also in various meteorological operational fields such as storm identification and tracking, vertical structure analysis, and numerical forecast assimilation. Reconstructing a complete and accurate horizontal reflectivity factor field using dual polarization parameters provides a universal data foundation for diverse downstream applications, demonstrating its high importance and versatility.

[0006] Therefore, developing a method that can fully utilize the multi-parameter observation advantages of dual-polarization radar to effectively repair horizontal reflectivity data under beam blocking conditions with a single radar is of great significance for overcoming the limitations of terrain on radar observation and improving the monitoring capabilities of a single radar station in complex terrain environments. It can also provide practical and reliable technical support for disaster prevention and mitigation work in areas lacking radar network coverage.

[0007] Therefore, proposing a self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization to solve the difficulties existing in the prior art is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0008] In view of this, the present invention provides a self-filling method for single-unit dual-polarization radar data based on physical constraints and dynamic optimization, which is used to solve the core technical problem of missing effective observation data caused by beam obstruction due to terrain in existing meteorological radar technology when a single radar is in complex terrain.

[0009] To achieve the above objectives, the present invention provides the following technical solution: A self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization includes the following steps: S1. Obtain the raw observation data of the dual-polarization radar, preprocess the raw observation data, and calculate the beam blocking rate of each pixel by combining the digital elevation model and the radar beam propagation model. S2. Establish reliability criteria, evaluate the reliability of dual polarization parameters for each pixel to be filled, and select the optimal filling path for each pixel based on the evaluation results. S3, Layered execution core and auxiliary filling path fill data, when compared to differential phase When deemed reliable, based on - The core path for relationship construction; when compared to differential phase Unreliable but differential reflectivity When reliable, based on - Statistical relationships are used to construct auxiliary paths, and data imputation is performed hierarchically. S4. Use the variational method to fuse the data and construct a cost function. By minimizing the cost function, solve for the final, physically coherent reflectivity field after filling.

[0010] Optionally, the preprocessing in S1 includes: quality control of the acquired raw observation data, including noise filtering, velocity deblurring, and non-meteorological echo identification.

[0011] Optionally, S1 also includes: setting a threshold based on the beam blocking rate to divide the radar scanning area into a data gap area, a data doubt area, and an effective observation area.

[0012] Optionally, the specific details of calculating the beam blocking rate for each pixel in S1 are as follows: S11, Coordinate Transformation and Beam Path Calculation; S12. Terrain profile extraction and occlusion detection; S13, Partial blocking calculation.

[0013] Optionally, the specific details of coordinate transformation and beam path calculation in S11 are as follows: S111. Determine the altitude of the beam center axis. First, considering the effects of Earth's curvature and atmospheric refraction, a 4 / 3 Earth radius model based on standard atmospheric refraction is used to correct the radar beam propagation path. (Pixel count) The elevation of the beam center at the location The formula is as follows:

[0014] in, The elevation angle of the radar antenna. The actual radius of the Earth. This is the equivalent Earth radius coefficient, usually taken as 4 / 3. The height for installing the radar antenna, The straight-line distance from the radar antenna to the target object; S112, Coordinate Transformation pixels From polar coordinates The data is converted to geographic coordinates, then to a geocentric rectangular coordinate system, and geometric calculations are performed within the same framework as the DEM data.

[0015] Optionally, the specific content of terrain profile extraction and occlusion judgment in S12 is as follows: S121. Extract terrain profile Along the azimuth of the radar beam Starting from the radar position and extending to the maximum calculated distance, extract a terrain elevation profile from the DEM data. ; S122, Preliminary occlusion assessment The elevation of the beam center axis With topographic elevation profile For a certain distance, a comparison can be made. ,satisfy This indicates that the beam center axis has been blocked by the terrain. In this case, the blocking rate of the pixel is usually determined directly. .

[0016] Optionally, the specific details of some blocking calculations in S13 are as follows: S131. Gaussian beam energy distribution model: The energy of a radar beam is not uniformly distributed, but follows a Gaussian distribution. On the cross-section, the power distribution function is:

[0017] in, It is the beam radius. ; The horizontal coordinate on the cross section The longitudinal coordinate on the cross-section; S132. Calculate beam blocking rate

[0018] Beam blocking rate The calculation formula is:

[0019] For the Gaussian beam energy distribution model, the calculation formula is as follows:

[0020] in, The intrusion depth of the terrain relative to the bottom of the beam, i.e. If the terrain is lower than the bottom of the beam, then , Current distance Beam radius at the location; S133, for each The above process is repeated for each pixel to generate a beam jamming rate spatial distribution map that perfectly matches the radar scanning grid. .

[0021] Optionally, the specific content of establishing the reliability criterion in S2 is as follows: Reliability criteria:

[0022] in, , , These are the thresholds for the correlation coefficient, signal-to-noise ratio, and differential phase gradient, respectively. Reliability criteria:

[0023] in, The polarization correlation coefficient represents the correlation between the horizontal and vertical echo signals. The reflectivity threshold, This is the azimuth direction index. This is the distance direction index.

[0024] Optionally, the formula for the cost function in S4 is as follows:

[0025] in, This is the original observation field; the missing values ​​are default values. , , These are binary masks for effective observations, core filling paths, and auxiliary filling paths, respectively. , , This is a weighting coefficient, which is proportional to the reliability of each data source; The smoothing constraint weights are used to control the smoothness of the final result. For the final, physically coherent reflectivity field, To fill the reflectivity field of the core path, To help fill in the reflectivity field of the path.

[0026] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for self-filling data of a single-part dual-polarization radar based on physical constraints and dynamic optimization, the beneficial effects of which are: 1) It relies solely on data from a single dual-polarization radar, without depending on expensive external radar networks or numerical models, making it widely applicable and cost-effective. 2) Filling is done using parameters such as KDP that are insensitive to blocking, ensuring that the recovered data has clear physical meaning rather than being purely mathematical interpolation; 3) The final spatial optimal fusion step ensures a smooth transition between the filled area and the surrounding effective observation area, forming a physically coherent and spatially consistent high-quality data field, laying a reliable foundation for subsequent applications such as quantitative precipitation estimation. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0028] Figure 1 The flowchart illustrates a self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization, as provided in this invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] This invention aims to solve the core technical challenge of existing meteorological radar technology, which suffers from the loss of effective observation data due to beam obstruction caused by terrain in single-radar coverage environments located in complex terrain conditions. The specific details are as follows: ① Solve the problem of data failure of a single radar in complex terrain: In complex terrain areas such as mountainous and hilly regions, the radar beam propagation path is blocked by mountains and other topographical features, resulting in a severe beam jamming effect. This jamming leads to a significant underestimation or complete absence of the horizontal reflectivity factor (ZH), a fundamental radar observation product, in the blocked areas, creating large-scale monitoring blind spots. Traditional interpolation methods, based solely on geometric relationships and lacking physical basis, cannot accurately recover the true atmospheric echo information of the blocked areas. This invention, based on the observation characteristics of dual-polarization radar, designs a system that uses parameters with stronger anti-jamming capabilities to repair and fill in the easily missing ZH parameters, enabling the use of a single dual-polarization radar to fill in missing data.

[0031] ② Addressing the geographical limitations of multi-radar composite technology: The current mainstream solution to beam jamming is multi-radar data fusion technology, but this method relies on a dense radar observation network. In practical applications, due to limitations such as construction costs, terrain conditions, and maintenance resources, many areas, especially remote mountainous areas and border regions, can only rely on a single radar for monitoring. In these areas lacking radar network coverage, multi-radar fusion technology is simply not feasible, resulting in the long-standing ineffective solution to the beam jamming problem. This invention addresses the issue of scarce radar network deployment by providing data supplementation specifically for single meteorological radars.

[0032] ③ Solve the problem of insufficient utilization of dual polarization parameters: While modern dual-polarization radars can provide multiple observation parameters, including differential phase ratio (KDP) and differential reflectivity (ZDR), and these parameters are inherently resistant to beam jamming, existing technologies lack effective mechanisms to fully utilize the complementary characteristics of these parameters. Particularly in the case of a single radar, a systematic methodology for repairing low-quality parameters using high-quality parameters has not yet been established. This invention identifies highly reliable dual-polarization parameters for different environmental regions and adopts targeted solutions at different levels, fully utilizing different dual-polarization observation parameters to improve the accuracy and reliability of radar data filling.

[0033] See Figure 1 As shown, this invention discloses a self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization, comprising the following steps: S1. Acquire the raw observation data of the dual-polarization radar, preprocess the raw observation data, and calculate the beam blocking rate of each pixel by combining the digital elevation model and the radar beam propagation model (including geometric propagation and energy distribution). S2. Establish reliability criteria, evaluate the reliability of dual polarization parameters for each pixel to be filled, and select the optimal filling path for each pixel based on the evaluation results. S3, Layered execution core and auxiliary filling path fill data, when compared to differential phase When deemed reliable, based on - The core path for relationship construction; when compared to differential phase Unreliable but differential reflectivity When reliable, based on - Statistical relationships are used to construct auxiliary paths, and data imputation is performed hierarchically. Specifically, the core filling path: based on - Physical relationship model: when When =1, this path is started; Local relation establishment: At the edge of the blocked region, select an unblocked learning window W containing N pixels. Using the data within the window, fit the local relation through robust regression.

[0034] Among them, coefficient and By minimizing the objective function get.

[0035] Data reconstruction: Reliable data within the congested area Substituting the values ​​into the above relationship, the reconstructed reflectance is calculated:

[0036] Auxiliary filling path: based on - Statistical relationship model: when =0 but When this path is activated, it will be started.

[0037] Establishing local relationships: Also within the learning window W, establish... and Linear statistical relationship:

[0038] Data reconstruction: utilizing reliable methods within the congestion zone Make an estimate:

[0039] in, For the slope parameter, This is the intercept parameter.

[0040] S4. Use the variational method to fuse the data and construct a cost function. By minimizing the cost function, solve for the final, physically coherent reflectivity field after filling.

[0041] Furthermore, the preprocessing in S1 includes quality control of the acquired raw observation data, including noise filtering, velocity deblurring, and non-meteorological echo identification.

[0042] Furthermore, S1 also includes: setting a threshold based on the beam blocking rate to divide the radar scanning area into a data gap area, a data doubt area, and an effective observation area.

[0043] Furthermore, the specific details of calculating the beam blocking rate for each pixel in S1 are as follows: S11, Coordinate Transformation and Beam Path Calculation; S12. Terrain profile extraction and occlusion detection; S13, Partial blocking calculation.

[0044] Furthermore, the specific details of coordinate transformation and beam path calculation in S11 are as follows: S111. Determine the altitude of the beam center axis. First, considering the effects of Earth's curvature and atmospheric refraction, a 4 / 3 Earth radius model based on standard atmospheric refraction is used to correct the radar beam propagation path. (Pixel count) The elevation of the beam center at the location The formula is as follows:

[0045] in, The elevation angle of the radar antenna. The actual radius of the Earth. This is the equivalent Earth radius coefficient, usually taken as 4 / 3. The height for installing the radar antenna, The straight-line distance from the radar antenna to the target object; S112, Coordinate Transformation pixels From polar coordinates The data is converted to geographic coordinates, then to a geocentric rectangular coordinate system, and geometric calculations are performed within the same framework as the DEM data.

[0046] Furthermore, the specific details of terrain profile extraction and occlusion determination in S12 are as follows: S121. Extract terrain profile Along the azimuth of the radar beam Starting from the radar position and extending to the maximum calculated distance, extract a terrain elevation profile from the DEM data. ; S122, Preliminary occlusion assessment The elevation of the beam center axis With topographic elevation profile For a certain distance, a comparison can be made. ,satisfy This indicates that the beam center axis has been blocked by the terrain. In this case, the blocking rate of the pixel is usually determined directly. .

[0047] Furthermore, the specific details of the blocking calculations in S13 are as follows: S131. Gaussian beam energy distribution model: The energy of a radar beam is not uniformly distributed, but follows a Gaussian distribution. On the cross-section, the power distribution function is:

[0048] in, It is the beam radius. ; The horizontal coordinate on the cross section The longitudinal coordinate on the cross-section; S132. Calculate beam blocking rate

[0049] Beam blocking rate The calculation formula is:

[0050] Specifically, beam blocking rate Defined as the percentage of beam energy obscured by terrain relative to the total energy. This is an integral calculation performed over the beam cross-section. When the bottom of the beam is raised by the terrain (the raised height is...) When blocked, its beam blocking rate The formula for calculating is as shown above.

[0051] For the Gaussian beam energy distribution model, the calculation formula is as follows:

[0052] in, The intrusion depth of the terrain relative to the bottom of the beam, i.e. If the terrain is lower than the bottom of the beam, then , Current distance Beam radius at the location; S133, for each The above process is repeated for each pixel to generate a beam jamming rate spatial distribution map that perfectly matches the radar scanning grid. .

[0053] Furthermore, the specific details of establishing reliability criteria in S2 are as follows: Reliability criteria:

[0054] in, , , These are the thresholds for the correlation coefficient, signal-to-noise ratio, and differential phase gradient, respectively. Reliability criteria:

[0055] in, The polarization correlation coefficient represents the correlation between the horizontal and vertical echo signals. The reflectivity threshold, This is the azimuth direction index. This is the distance direction index.

[0056] Furthermore, the formula for the cost function in S4 is as follows:

[0057] in, This is the original observation field; the missing values ​​are default values. , , These are binary masks for effective observations, core filling paths, and auxiliary filling paths, respectively. , , This is a weighting coefficient, which is proportional to the reliability of each data source; The smoothing constraint weights are used to control the smoothness of the final result. For the final, physically coherent reflectivity field, To fill the reflectivity field of the core path, To help fill in the reflectivity field of the path.

[0058] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0059] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization, characterized in that, Includes the following steps: S1. Obtain the raw observation data of the dual-polarization radar, preprocess the raw observation data, and calculate the beam blocking rate of each pixel by combining the digital elevation model and the radar beam propagation model. S2. Establish reliability criteria, evaluate the reliability of dual polarization parameters for each pixel to be filled, and select the optimal filling path for each pixel based on the evaluation results. S3, Layered execution core and auxiliary filling path fill data, when compared to differential phase When deemed reliable, based on - Core path to relationship building; When the differential phase Unreliable but differential reflectivity When reliable, based on - Statistical relationships are used to construct auxiliary paths, and data imputation is performed hierarchically. S4. Use the variational method to fuse the data and construct a cost function. By minimizing the cost function, solve for the final, physically coherent reflectivity field after filling.

2. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 1, characterized in that, The preprocessing in S1 involves quality control of the acquired raw observation data, including noise filtering, velocity deblurring, and non-meteorological echo identification.

3. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 1, characterized in that, S1 also includes: setting a threshold based on the beam blocking rate to divide the radar scanning area into a data gap area, a data doubt area, and an effective observation area.

4. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 1, characterized in that, The specific steps for calculating the beam blocking rate for each pixel in S1 are as follows: S11, Coordinate Transformation and Beam Path Calculation; S12. Terrain profile extraction and occlusion detection; S13, Partial blocking calculation.

5. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 4, characterized in that, The specific details of coordinate transformation and beam path calculation in S11 are as follows: S111. Determine the altitude of the beam center axis. First, considering the effects of Earth's curvature and atmospheric refraction, a 4 / 3 Earth radius model based on standard atmospheric refraction is used to correct the radar beam propagation path. (Pixel count) The elevation of the beam center at the location The formula is as follows: in, The elevation angle of the radar antenna. The actual radius of the Earth. This is the equivalent Earth radius coefficient, usually taken as 4 / 3. The height for installing the radar antenna, The straight-line distance from the radar antenna to the target object; S112, Coordinate Transformation pixels From polar coordinates The data is converted to geographic coordinates, then to a geocentric rectangular coordinate system, and geometric calculations are performed within the same framework as the DEM data.

6. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 4, characterized in that, The specific content of terrain profile extraction and occlusion judgment in S12 is as follows: S121. Extract terrain profile Along the azimuth of the radar beam Starting from the radar position and extending to the maximum calculated distance, extract a terrain elevation profile from the DEM data. ; S122, Preliminary occlusion assessment The elevation of the beam center axis With topographic elevation profile For a certain distance, a comparison can be made. ,satisfy This indicates that the beam center axis has been blocked by the terrain. In this case, the blocking rate of the pixel is usually determined directly. .

7. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 4, characterized in that, The specific details of the blocking calculations in S13 are as follows: S131. Gaussian beam energy distribution model: The energy of a radar beam is not uniformly distributed, but follows a Gaussian distribution. On the cross-section, the power distribution function is: in, It is the beam radius. ; The horizontal coordinate on the cross section The longitudinal coordinate on the cross-section; S132. Calculate beam blocking rate Beam blocking rate The calculation formula is: For the Gaussian beam energy distribution model, the calculation formula is as follows: in, The intrusion depth of the terrain relative to the bottom of the beam, i.e. If the terrain is lower than the bottom of the beam, then , Current distance Beam radius at the location; S133, for each The above process is repeated for each pixel to generate a beam jamming rate spatial distribution map that perfectly matches the radar scanning grid. .

8. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 1, characterized in that, The specific content of the reliability criterion established in S2 is as follows: Reliability criteria: in, , , These are the thresholds for the correlation coefficient, signal-to-noise ratio, and differential phase gradient, respectively. Reliability criteria: in, The polarization correlation coefficient represents the correlation between the horizontal and vertical echo signals. The reflectivity threshold, This is the azimuth direction index. This is the distance direction index.

9. The self-filling method for single-part dual-polarization radar data based on physical constraints and dynamic optimization according to claim 1, characterized in that, The formula for the cost function in S4 is as follows: in, This is the original observation field; the missing values ​​are default values. , , These are binary masks for effective observations, core filling paths, and auxiliary filling paths, respectively. , , This is a weighting coefficient, which is proportional to the reliability of each data source; The smoothing constraint weights are used to control the smoothness of the final result. For the final, physically coherent reflectivity field, To fill the reflectivity field of the core path, To help fill in the reflectivity field of the path.

Images