A single-station ultra-high frequency radar river cross-section surface flow velocity inversion method based on RAMPJI

Abstract

This invention discloses a single-station UHF radar method for inverting surface velocity at river cross sections based on RAMPJI. The method includes acquiring radar echo data, dividing the observed flow field area into multiple equally spaced intervals according to the cross-sectional direction, estimating the main flow direction, back-projecting each radial velocity component to the main flow direction to construct a candidate set of velocity values ​​and eliminate outliers, then introducing a weighted fusion mechanism for each radial velocity angle pair within each partition to obtain stable observation values, and employing a Top-K strategy to weightedly fuse effective angle pairs to achieve cross-sectional surface velocity inversion. This invention achieves effective fusion of multi-angle radial velocity observation information and inversion of cross-sectional surface velocity. It can stably invert cross-sectional surface velocity under different environmental noise, different flow direction offsets, and missing measurement conditions. It maintains high inversion accuracy and spatial continuity in complex river environments, and its cross-sectional surface velocity results show good consistency with the velocity values ​​measured by an Acoustic Doppler Current Profiler (ADCP).

Description

Technical Field

[0001] This invention belongs to the field of ultra-high frequency radar detection of river hydrology, specifically involving a single-station ultra-high frequency radar surface velocity inversion method based on RAMPJI. Background Technology

[0002] Ultra-high frequency (UHF) radar offers advantages such as non-contact operation, all-weather capability, and wide coverage. Because its operating frequency band matches the roughness scale of the river surface, it can effectively acquire information from the scattered echoes from the river surface, gradually becoming an important means of detecting river surface velocity. Radar measures flow by inverting the surface velocity of the river cross-section through the acquisition of electromagnetic scattered echo information from the river surface, avoiding the safety risks and operational limitations associated with direct contact between the instrument and the water body in traditional measurement methods. It is particularly suitable for long-term continuous observation under conditions of wide channels, high flow velocities, and floods.

[0003] Existing UHF radar flow measurement modes mainly include two types: multi-station and single-station. The multi-station mode deploys multiple radars to simultaneously acquire radial velocities at different azimuth angles and synthesizes vector velocities, which can effectively alleviate the problem of decreased inversion accuracy caused by the radial velocity approaching zero near the normal direction in the single-station mode (i.e., the "zero-velocity zone" problem). However, the multi-station mode is constrained by factors such as terrain deployment conditions, equipment costs, and requirements for synchronization and data integrity, making its actual deployment and promotion quite difficult.

[0004] In contrast, the single-station model relies solely on radial velocity observations from a single station to achieve cross-sectional surface velocity inversion, offering advantages such as flexible deployment, system simplicity, and ease of maintenance. Several methods for extracting cross-sectional surface velocity have been developed around the single-station model. The single-point projection method, assuming a fixed and known main flow direction, projects the radial velocity onto the main flow direction to obtain a surface velocity estimate. However, in natural rivers, the flow direction often changes with time and space, and a normal zero-velocity region exists, making the results prone to distortion and sensitive to noise and outliers. The Teague method avoids the influence of the normal zero-velocity region by selecting symmetrical observations within a certain angular domain on both sides of the radar normal, converting horizontal velocity using geometric relationships, and averaging the results. However, in the non-uniform flow field of natural rivers ("lower near the bank, higher in the middle"), the fan-shaped distance gate may simultaneously cover different cross-sectional locations, leading to estimation bias. Furthermore, this method is highly dependent on data integrity; missing measurements reduce stability. The surface projection method further expands the set of observations involved in the estimation, using the median of all velocity observations within each distance cell as the velocity estimate at that location to enhance adaptability to complex flow fields. However, when the angular domain is too large, it is easy to introduce radial components that are unrelated to the cross-sectional velocity. In addition, the data quality is poor in the low radial velocity region near the normal and in the interference region of near-shore fixed scatterers. In engineering applications, it is usually necessary to remove low-quality angular domains and retain only the effective angular range for calculation. Summary of the Invention

[0005] To address the aforementioned problems, the present invention aims to provide a single-station UHF radar method for retrieving surface flow velocity from river cross sections based on RAMPJI (Regional Adaptive Multi-Angle Pairing Joint Inversion).

[0006] The specific technical solution for achieving the objective of this invention is as follows:

[0007] A method for inverting surface flow velocity in a river cross-section using a single-station UHF radar based on RAMPJI includes the following steps:

[0008] Step 1: Acquire and process multi-channel echo data of the river surface from a single-station UHF side-scan radar to obtain radial velocity observation results at each range gate and azimuth angle;

[0009] Step 2: Divide the observation flow field region into multiple equally spaced intervals according to the cross-sectional direction, with each interval corresponding to an independent cross-sectional sub-region;

[0010] Step 3: Estimate the main flow direction of the river using the full-angle energy minimization criterion;

[0011] Step 4: Based on the geometric projection relationship, back-project each radial velocity component to the main flow direction to construct a candidate set of velocity values;

[0012] Step 5: Identify and eliminate outliers based on the MAD criterion;

[0013] Step 6: Introduce a weighted fusion mechanism for each radial velocity angle pair within each partition, and assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values;

[0014] Step 7: Use the Top-K strategy to perform weighted fusion of effective angle pairs to achieve cross-sectional surface velocity inversion.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0016] (1) This invention addresses the problem that the inversion of cross-sectional surface velocity by single-station UHF radar under complex river conditions is easily affected by changes in flow direction, near-shore interference, noise, and outliers. It constructs a complete inversion process that includes longitudinal partitioning of the cross-section, adaptive estimation of the main flow direction, grid-by-grid back projection, robust pre-screening based on MAD, and multi-angle weighted fusion, realizing the effective fusion of radial velocity observation information from multiple angles and the inversion of cross-sectional surface velocity. This method can stably invert cross-sectional surface velocity under different environmental noise, different flow direction offsets, and missing measurement conditions. It can maintain high inversion accuracy and spatial continuity in complex river environments. Its cross-sectional surface velocity results have good consistency with the velocity values ​​measured by the Acoustic Doppler Current Profiler (ADCP).

[0017] (2) The single-station UHF radar river cross-section surface velocity inversion method based on regional adaptive multi-angle pair joint inversion proposed in this invention fully exploits the radial velocity observation information of multi-angle pairs under single-station conditions. It realizes the cross-section surface velocity inversion through longitudinal partitioning of the cross-section and weighted fusion of multi-angle pairs, avoiding the dependence on multi-station collaborative deployment and strict time synchronization. The system structure is simpler, and the engineering implementation difficulty and cost are significantly reduced.

[0018] (3) The method of this invention breaks through the dependence of traditional single-point projection method and fixed angle domain method on flow direction stability and finite symmetry angle pairs. By jointly participating in and adaptively filtering multi-angle pair observation data, it effectively avoids the adverse effects of the normal zero velocity region on the inversion accuracy. Even when the flow direction changes significantly or the observation angle domain is incomplete, it can still maintain stable and reliable cross-sectional surface velocity estimation results.

[0019] (4) In view of the non-uniform spatial distribution of flow velocity in natural river cross sections, the present invention introduces a partitioning strategy, which enables the radial flow velocity observation at different radial distances and different azimuth angles to more reasonably characterize the flow velocity characteristics at each position of the cross section, reduces the systematic deviation caused by the fan-shaped distance gate covering different cross section positions, and improves the adaptability to complex river flow fields.

[0020] (5) In the process of multi-angle radial velocity observation fusion, the present invention introduces a robust inversion mechanism based on MAD criterion and a multi-weight function fusion mechanism, which can effectively suppress the influence of adverse factors such as noise interference, outliers and near-shore fixed scatterers, reduce the influence of individual low-quality observations on the overall inversion results, and improve the stability, reliability and engineering applicability of cross-sectional surface velocity inversion.

[0021] The present invention will be further described below with reference to specific embodiments. Attached Figure Description

[0022] Figure 1This is a flowchart of the single-station UHF radar surface velocity inversion method for river cross-sections based on RAMPJI, according to the present invention.

[0023] Figure 2 This is a schematic diagram illustrating the implementation of the flow velocity inversion method of the present invention.

[0024] Figure 3 This is a schematic diagram of the radial velocity components back-projected onto the main flow direction in an embodiment of the present invention.

[0025] Figure 4 This is a comparison chart of the surface velocity inversion results for a single cross-section in an embodiment of the present invention.

[0026] Figure 5 This is a comparison chart of the surface velocity inversion results of three methods and ADCP data curves at a cross-section of 150 meters on May 7, 2025, in an embodiment of the present invention.

[0027] Figure 6 This is a scatter plot comparing the surface velocity inversion results of three methods with ADCP data at a cross-section of 150 meters on May 7, 2025, in an embodiment of the present invention. Detailed Implementation

[0028] Example

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0031] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0032] Combination Figure 1 and Figure 2 A method for inverting surface flow velocity in a river cross-section using a single-station UHF radar based on RAMPJI includes the following steps:

[0033] Step 1: Acquire multi-channel echo data of the river surface from a single-station UHF side-scan radar. Process the echo data in the range, Doppler, and angular dimensions to obtain the radial velocity observation results at each range gate and azimuth angle.

[0034] Step 1-1: Record the number of antenna channels as... The number of distance units is The number of frames is Generate a dataset containing all antenna channels and all frames. Distance dimension spectral data, denoted as For each antenna channel, windowing and the first time for each frame of data are performed. (Fast Fourier Transform) yields :

[0035]

[0036] in, , For a certain antenna channel, For a certain data frame, For all distance units, For Fast Fourier Transform, To be The zero-frequency component in the result has shifted to the center position. For Hanming window;

[0037] Steps 1-2: For the data matrix For each antenna channel, windowing and a second time are performed on each range element. ,get :

[0038]

[0039] in, , For a certain distance unit, For all frames;

[0040] Steps 1-3: Construct a spatial covariance matrix for the multi-channel echo data in each range cell, and use the Multiple Signal Classification (MUSIC) algorithm based on eigenvalue decomposition to estimate the high-resolution angle of arrival of the signal and obtain the azimuth information of each scattering surface element.

[0041] Steps 1-4: Based on the obtained range, Doppler, and angular data, convert the Doppler frequency shift into radial velocity observations at the corresponding range gates and azimuth angles. This yields a set of radial velocity observation results covering multiple range gates and observation azimuths of the river cross-section, providing basic observation data for subsequent cross-sectional surface velocity calculations.

[0042] Step 2: Divide the observation flow field region into multiple equally spaced intervals along the cross-sectional direction, with each interval corresponding to an independent cross-sectional sub-region:

[0043] The observation grid is divided into multiple equally spaced intervals according to the cross-sectional direction, with each interval corresponding to an independent cross-sectional sub-region. The range resolution of the reference radar for each interval is selected. This longitudinal partitioning method can effectively reduce the interference of nearshore velocity components on the inversion of surface velocity at long-distance gate sections, thus making the inversion results more consistent with the spatial distribution characteristics of the actual flow field.

[0044] Step 3: Estimate the main flow direction of the river using the full-angle energy minimization criterion:

[0045] Step 3-1: When the radar observation direction is perpendicular to the actual mainstream direction, the sum of the obtained radial velocity components is minimized. By searching for the angle with the minimum energy, the adaptive determination of the mainstream direction can be achieved. This process iterates through every candidate angle across the entire flow field. Calculate the cumulative energy of the absolute values ​​of all radial velocities in that direction. By finding ways to accumulate energy and Angle that achieves minimum value This allows us to determine the radar observation azimuth angle perpendicular to the mainstream direction:

[0046]

[0047]

[0048] in, Indicates radial distance Location, observation azimuth angle is radial flow velocity;

[0049] Step 3-2: Based on the radar observation azimuth angle perpendicular to the main current direction obtained in Step 3-1. The mainstream direction angle can be obtained. for:

[0050]

[0051] Step 4: Based on the geometric projection relationship, backproject each radial velocity component onto the main flow direction, such as... Figure 3 As shown, a set of candidate values ​​for the flow rate is constructed:

[0052] For any grid point Its back projection onto the upward component of the mainstream is:

[0053]

[0054] Radial velocities at different angles were uniformly mapped to the mainstream direction, providing crucial multi-angle redundant observation data for subsequent robust screening and weighted fusion.

[0055] Step 5: Identify and exclude outliers based on the MAD criterion:

[0056] Step 5-1: For each distance interval According to its back-projected velocity set , This represents the total number of available observation angles within the interval, and the median is calculated. and value:

[0057]

[0058] Step 5-2: Construct an adaptive threshold:

[0059]

[0060] in, This is a multiple of the threshold (generally taken as 1 to 3). The smaller the size, the stricter the screening process. These are Gaussian distribution correction coefficients;

[0061] For interval A certain back-projected flow velocity value within If it satisfies If a point is found to be a reliable observation, it is retained; otherwise, it is removed as an outlier. This significantly enhances the robustness of the inversion algorithm against outliers and local disturbances, thus providing high-quality data input for subsequent weighted fusion.

[0062] Step 6: Within each partition, for each radial velocity angle pair corresponding to the outlier-excluded flow velocity, a weighted fusion mechanism is introduced to assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values.

[0063] Step 6-1: Introduce geometric degradation suppression weights to adaptively weaken unreliable observations from near the zero-velocity region, and define corner pairs. Angle with the center of the zero speed zone nearest angular distance:

[0064]

[0065] Based on this, Tukey's two weights are introduced, and a geometric degradation suppression weight function is defined. :

[0066]

[0067] in, The half-width at half maximum (WWHM) of the suppression region is generally taken as... , As the lowest weight, it is usually set to 0.05 to 0.15. That is, when one of the selected angle pairs falls into the zero velocity region, the weight takes the minimum value. This weighting function can adaptively weaken unreliable observations from near the zero velocity region, thereby effectively improving the stability of multi-angle fusion.

[0068] Step 6-2: Introduce amplitude reliability weights to enhance the contribution of high-amplitude observations in the fusion. For a symmetrical angle pair, its two radial velocity observations... and The reliability is determined by the smaller of the absolute values ​​of the amplitudes, that is:

[0069]

[0070] Define the amplitude reliability weighting function as follows:

[0071]

[0072] in, To control the saturation intensity (typically taken as 0.2 m / s), the amplitude reliability weighting function can adaptively enhance the contribution of high-amplitude, high-stability observation points, while effectively suppressing the adverse effects of small-amplitude observations near the normal, thereby significantly enhancing the geometrically well-posedness of velocity inversion and the stability of the overall results.

[0073] Step 6-3: Introduce symmetry consistency weights to evaluate the reliability of the match between symmetry angle pairs and observed data. Ideally, the radial velocity amplitudes of symmetry angle pairs should be similar; excessive differences indicate interference. The weights are defined as follows:

[0074]

[0075] in, To avoid extremely small positive numbers with a denominator of zero;

[0076] The symmetry consistency weight is between 0 and 1: when the radial velocity amplitudes on both sides are close, the weight approaches 1, indicating high consistency and high reliability of the angle pair; when the amplitude difference increases, the weight automatically decreases, thereby suppressing the influence of abnormal angle pairs.

[0077] Step 6-4: Determine the overall mass weight for each radial velocity angle pair as follows:

[0078]

[0079] Step 7: Use the Top-K strategy to perform weighted fusion of effective angle pairs to achieve cross-sectional surface velocity inversion.

[0080] Let the effective set of observation angles be... , its elements Corresponding comprehensive weight Sort in descending order, keeping only the first few. The angles corresponding to the maximum weights constitute a subset. ;

[0081] Based on this subset, a weighted fusion is performed to obtain the distance interval. Inverted values ​​of horizontal flow velocity within:

[0082]

[0083] in, For elements Its corresponding comprehensive weight.

[0084] This Top-K weighted fusion strategy selects the most reliable observation angles for calculation, which effectively suppresses the influence of abnormal angles and zero-velocity regions while ensuring geometric well-posedness, and finally obtains a stable, smooth and physically consistent cross-sectional surface velocity distribution.

[0085] Figure 4 This is a comparison chart of the surface velocity inversion results of a single cross-section in a certain embodiment scenario. This chart preliminarily proves the feasibility of inverting the surface velocity of a river cross-section using the RAMPJI algorithm.

[0086] Figure 5 and Figure 6 The results show comparisons of surface velocity inversion using three methods with ADCP data curves and scatter plots at a cross-section of 150 meters on May 7, 2025. The experimental river section, characterized by variable flow direction, complex boundaries, and low signal-to-noise ratio, provided a rigorous testing environment for the algorithm. The comparative results demonstrate that RAMPJI is highly consistent with ADCP velocity data in both single-field observations and long-term sequences, with a significantly lower RMSE compared to the Teague and surface projection methods. Furthermore, RAMPJI maintains flow field continuity and physical plausibility even under conditions of flow direction deviation, boundary interference, and missing observations, showcasing its potential for long-term automated observation in complex river environments.

[0087] Overall, the RAMPJI velocity inversion method proposed in this scheme has significant advantages in terms of accuracy, robustness and generalizability. Its partitioned adaptive and multi-angle weighted fusion mechanism provides a general and scalable framework for non-contact velocity inversion.

[0088] The verification in this embodiment confirms that the method of the present invention is effective and advantageous in the inversion of surface flow velocity at river cross-sections.

[0089] This solution also provides a single-station UHF radar surface velocity inversion system for river cross-sections based on RAMPJI, which includes the following modules:

[0090] Data processing module: used to acquire and process multi-channel echo data of the river surface from a single-station UHF side-scan radar, and obtain radial velocity observation results at each range gate and azimuth angle;

[0091] The observation flow field region is divided into multiple equally spaced intervals according to the cross-sectional direction, and each interval corresponds to an independent cross-sectional sub-region.

[0092] River mainstream direction estimation module: used to estimate the river mainstream direction using the full-angle energy minimization criterion;

[0093] Effective angle determination module: It is used to back-project each radial velocity component to the main flow direction based on the geometric projection relationship, so as to construct a candidate value set of velocity and identify and eliminate outliers based on the MAD criterion;

[0094] Velocity inversion module: It is used to introduce a weighted fusion mechanism for each radial velocity angle pair in each partition, and assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values; the Top-K strategy is used to perform weighted fusion of effective angle pairs to realize the inversion of cross-sectional surface velocity.

[0095] This solution also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0096] Step 1: Acquire and process multi-channel echo data of the river surface from a single-station UHF side-scan radar to obtain radial velocity observation results at each range gate and azimuth angle;

[0097] Step 2: Divide the observation flow field region into multiple equally spaced intervals according to the cross-sectional direction, with each interval corresponding to an independent cross-sectional sub-region;

[0098] Step 3: Estimate the main flow direction of the river using the full-angle energy minimization criterion;

[0099] Step 4: Based on the geometric projection relationship, back-project each radial velocity component to the main flow direction to construct a candidate set of velocity values;

[0100] Step 5: Identify and eliminate outliers based on the MAD criterion;

[0101] Step 6: Introduce a weighted fusion mechanism for each radial velocity angle pair within each partition, and assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values;

[0102] Step 7: Use the Top-K strategy to perform weighted fusion of effective angle pairs to achieve cross-sectional surface velocity inversion.

[0103] This solution also provides a computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, performs the following steps:

[0104] Step 1: Acquire and process multi-channel echo data of the river surface from a single-station UHF side-scan radar to obtain radial velocity observation results at each range gate and azimuth angle;

[0105] Step 2: Divide the observation flow field region into multiple equally spaced intervals according to the cross-sectional direction, with each interval corresponding to an independent cross-sectional sub-region;

[0106] Step 3: Estimate the main flow direction of the river using the full-angle energy minimization criterion;

[0107] Step 4: Based on the geometric projection relationship, back-project each radial velocity component to the main flow direction to construct a candidate set of velocity values;

[0108] Step 5: Identify and eliminate outliers based on the MAD criterion;

[0109] Step 6: Introduce a weighted fusion mechanism for each radial velocity angle pair within each partition, and assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values;

[0110] Step 7: Use the Top-K strategy to perform weighted fusion of effective angle pairs to achieve cross-sectional surface velocity inversion.

[0111] The embodiments described above are merely one implementation method of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for inverting surface flow velocity in a river cross-section using a single-station UHF radar based on RAMPJI, characterized in that, Includes the following steps: Step 1: Acquire and process multi-channel echo data of the river surface from a single-station UHF side-scan radar to obtain radial velocity observation results at each range gate and azimuth angle; Step 2: Divide the observation flow field region into multiple equally spaced intervals according to the cross-sectional direction, with each interval corresponding to an independent cross-sectional sub-region; Step 3: Estimate the main flow direction of the river using the full-angle energy minimization criterion; Step 4: Based on the geometric projection relationship, back-project each radial velocity component to the main flow direction to construct a candidate set of velocity values; Step 5: Identify and eliminate outliers based on the MAD criterion; Step 6: Introduce a weighted fusion mechanism for each radial velocity angle pair within each partition, and assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values; Step 7: Use the Top-K strategy to perform weighted fusion of effective angle pairs to achieve cross-sectional surface velocity inversion.

2. The method for inverting surface flow velocity of a river cross-section based on RAMPJI using a single-station ultra-high frequency radar according to claim 1, characterized in that, The acquisition of radial velocity observation results at each distance gate and each azimuth angle in step 1 is specifically as follows: Step 1-1: Record the number of antenna channels as... The number of distance units is The number of frames is Generate a dataset containing all antenna channels and all frames. Distance dimension spectral data, denoted as For each antenna channel, windowing and the first time for each frame of data are performed. ,get : ； in, , For a certain antenna channel, For a certain data frame, For all distance units, For Fast Fourier Transform, To be The zero-frequency component in the result has shifted to the center position. For Hanming window; Steps 1-2: For the data matrix For each antenna channel, windowing and a second time are performed on each range element. ,get : ； in, , For a certain distance unit, For all frames; Steps 1-3: Construct a spatial covariance matrix for the multi-channel echo data within each range cell, and use the MUSIC algorithm based on eigenvalue decomposition to perform high-resolution angle of arrival estimation for the signal, thereby obtaining the azimuth information of each scattering surface element. Steps 1-4: Based on the obtained range, Doppler, and angular data, convert the Doppler frequency shift into radial velocity observations at the corresponding range gates and azimuth angles, thereby obtaining a set of radial velocity observation results covering multiple range gates and multiple observation azimuths of the river cross section.

3. The method for inverting surface flow velocity of a river cross-section based on RAMPJI using a single-station ultra-high frequency radar according to claim 1, characterized in that, The estimation of the main river flow direction in step 3 is specifically as follows: Step 3-1: Traverse every candidate angle throughout the entire flow field. Calculate the cumulative energy of the absolute values ​​of all radial velocities in that direction. By finding ways to accumulate energy and Angle that achieves minimum value This allows us to determine the radar observation azimuth angle perpendicular to the mainstream direction: ； ； in, Indicates radial distance Location, observation azimuth angle is radial flow velocity; Step 3-2: Based on the radar observation azimuth angle perpendicular to the main current direction obtained in Step 3-1. The mainstream direction angle can be obtained. for: 。 4. The method for inverting surface flow velocity of a river cross-section based on RAMPJI using a single-station ultra-high frequency radar according to claim 1, characterized in that, The candidate set of flow velocity values ​​in step 4 is specifically as follows: For any grid point Its back projection onto the upward component of the mainstream is: ； Radial velocities at different angles are uniformly mapped to the mainstream direction.

5. The method for inverting surface flow velocity of a river cross-section based on RAMPJI using a single-station ultra-high frequency radar according to claim 4, characterized in that, The exclusion of outliers in step 5 specifically involves: Step 5-1: For each distance interval According to its back-projected velocity set , This represents the total number of available observation angles within the interval, and the median is calculated. and value: ； Step 5-2: Construct an adaptive threshold: ； in, Multiples of the threshold These are Gaussian distribution correction coefficients; For interval A certain back-projected flow velocity value within If it satisfies If a point is found to be a reliable observation, it will be retained; otherwise, it will be removed as an outlier.

6. The method for inverting surface flow velocity of a river cross-section based on RAMPJI using a single-station ultra-high frequency radar according to claim 1, characterized in that, In step 6, assigning a comprehensive quality weight to each radial velocity angle pair to obtain stable observations specifically involves: Step 6-1: Introduce geometric degradation suppression weights to adaptively weaken unreliable observations from near the zero-velocity region, and define corner pairs. Angle with the center of the zero speed zone nearest angular distance: ； Based on this, Tukey's two weights are introduced, and a geometric degradation suppression weight function is defined. : ； in, For the suppression region half-width, As the lowest weight, when That is, when one of the selected angle pairs falls into the zero velocity region, the weight takes the minimum value. ; Step 6-2: Introduce amplitude reliability weights. For a pair of symmetrical angles, the two radial velocity observations... and The reliability is determined by the smaller of the absolute values ​​of the amplitudes, that is: ； Define the amplitude reliability weighting function as follows: ； in, To control saturation intensity; Step 6-3: Introduce symmetry-consistent weights: ； in, To avoid extremely small positive numbers with a denominator of zero; Step 6-4: Determine the overall mass weight for each radial velocity angle pair as follows: 。 7. The method for inverting surface flow velocity of a river cross-section based on RAMPJI using a single-station ultra-high frequency radar according to claim 6, characterized in that, The weighted fusion of valid angle pairs in step 7 is specifically as follows: Let the effective set of observation angles be... , its elements Corresponding comprehensive weight Sort in descending order, keeping only the first few. The angles corresponding to the maximum weights constitute a subset. ; Based on this subset, a weighted fusion is performed to obtain the distance interval. Inverted values ​​of horizontal flow velocity within: ； in, For elements Its corresponding comprehensive weight.

8. A single-station UHF radar surface velocity inversion system for river cross-sections based on RAMPJI, characterized in that, Includes the following modules: Data processing module: used to acquire and process multi-channel echo data of the river surface from a single-station UHF side-scan radar, and obtain radial velocity observation results at each range gate and azimuth angle; The observation flow field region is divided into multiple equally spaced intervals according to the cross-sectional direction, and each interval corresponds to an independent cross-sectional sub-region. River mainstream direction estimation module: used to estimate the river mainstream direction using the full-angle energy minimization criterion; Effective angle determination module: It is used to back-project each radial velocity component to the main flow direction based on the geometric projection relationship, so as to construct a candidate value set of velocity and identify and eliminate outliers based on the MAD criterion; Velocity inversion module: It is used to introduce a weighted fusion mechanism for each radial velocity angle pair in each partition, and assign a comprehensive quality weight to each radial velocity angle pair to obtain stable observation values; the Top-K strategy is used to perform weighted fusion of effective angle pairs to realize the inversion of cross-sectional surface velocity.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

10. A computer-storable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images