Sand-containing moving bed channel shore-based radar-sonar fusion flow monitoring method and system

Abstract

The invention discloses a sand-containing moving bed channel shore-based radar-sonar fusion flow monitoring method and system. The shore-based radar measuring unit obtains water surface speed information and optional water level information near a section; the sonar measuring unit is used for arranging a measuring point array along the section to obtain multi-point water depth and bed surface elevation; and the dynamic section reconstruction module reconstructs and updates the area and the boundary of the water-passing section in real time. The fusion and correction module comprises a speed mapping sub-module, a moving bed disturbance estimation sub-module, a sand-containing disturbance estimation sub-module and a fusion filtering sub-module, self-adaptive correction is carried out on mapping from the surface layer speed to the section average flow velocity according to radar and sonar data quality, bed surface erosion and deposition and sand-containing disturbance intensity, joint estimation is carried out with the section area, and real-time flow and quality indexes are output. The scheme does not depend on a fixed calibration curve, can adapt to rapid change of the section, and improves the stability and accuracy of continuous flow monitoring under the condition of high sand content.

Description

Technical Field

[0001] This invention relates to the field of irrigation canal hydrological monitoring and water resource metering technology, and in particular to a method and system for shore-based radar-sonar fusion flow monitoring suitable for sand-laden moving bed canals. Background Technology

[0002] Under conditions of high sediment content, irrigation canals are prone to rapid scouring and silting of the canal bed, leading to frequent changes in the cross-sectional shape. Traditional canal flow monitoring generally relies on a "water level-flow rate calibration curve" or a fixed cross-section assumption. When cross-sectional scouring and silting change, and roughness and hydraulic conditions vary with sediment content, the calibration curve quickly becomes invalid, resulting in increased measurement errors, high maintenance costs, and difficulty in achieving continuous and reliable real-time flow monitoring. Furthermore, single-sensor solutions have limitations: measuring only the water level is insufficient to reflect cross-sectional changes; measuring only the flow velocity, especially the surface velocity, makes it difficult to accurately infer the average cross-sectional velocity; and measuring only the cross-sectional geometry without velocity information also fails to obtain reliable flow rate. Therefore, there is an urgent need for a new method and system for real-time flow monitoring that can maintain stable accuracy under conditions of continuous changes in the moving bed cross-section and high sediment content disturbance. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for monitoring flow in a channel with a sandy dynamic bed using a fusion of shore-based radar and sonar. This method combines shore-based radar to measure water surface velocity and water level with sonar to observe dynamic cross-sections and bed scouring and silting. It is equipped with specific data fusion and correction algorithms to achieve real-time flow and quality indicators that do not rely on fixed calibration curves and can output under conditions of frequent cross-sectional changes.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: The flow monitoring method provided by this invention includes: a shore-based radar measurement unit acquiring water surface velocity and water level information; a sonar measurement unit acquiring dynamic cross-sectional geometry and bed elevation changes; quality control through a data acquisition and preprocessing module; updating the cross-sectional area and boundaries through a dynamic cross-section reconstruction module; introducing dynamic bed scouring and sedimentation disturbance indicators through a fusion and correction module to adaptively correct the mapping from water surface velocity to cross-sectional average flow velocity; and integrating or discretizing the velocity information and dynamic cross-section through a flow calculation module to output real-time flow and uncertainty and other quality indicators. The system provided by this invention includes at least a shore-based radar measurement unit, a sonar measurement unit, a time synchronization and positioning unit, a data acquisition and preprocessing module, a dynamic cross-section reconstruction module, a fusion and correction module, a flow calculation module, and a result output and communication module.

[0005] Beneficial effects: Compared with the prior art, the present invention has at least the following beneficial effects: 1. Resistance to Calibration Curve Failure: By combining sonar dynamic reconstruction of the cross-section with radar measurement of water surface velocity, the flow calculation does not depend on a fixed cross-section and a fixed calibration curve, adapting to cross-sectional changes caused by scouring and silting of the moving bed; 2. Strong adaptability to high sediment content: The fusion and correction module introduces sediment content disturbance index and moving bed disturbance term to adaptively correct the conversion of "surface velocity - cross-sectional average velocity" and reduce systematic deviation under high sediment content conditions; 3. Continuous real-time monitoring: Shore-based radar can achieve non-contact continuous observation, while sonar provides constraints on cross-sectional and bed surface changes. The two complement each other, improving the proportion and stability of available data. 4. Output quality is assessable: The system can simultaneously output quality indicators such as uncertainty, effective grid ratio, and fusion residual, which facilitates operation and maintenance and measurement reliability management. Attached Figure Description

[0006] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a block diagram of the overall structure of the system of the present invention; Figure 3 A schematic diagram of the deployment of shore-based radar and sonar at the cross-section of the channel. Figure 4 The algorithm functional block diagram for the fusion and correction module; In the diagram: 110, Shore-based radar measurement unit; 111, Radar antenna; 112, Radar ranging and velocity measurement processor; 120, Sonar measurement unit; 121, Sonar measuring point array; 122, Sonar controller; 130, Time synchronization and positioning unit; 150, Data acquisition and preprocessing module; 160, Dynamic cross-section reconstruction module; 170, Fusion and correction module; 171, Velocity mapping submodule; 172, Moving bed disturbance estimation submodule; 173, Sediment-laden disturbance estimation submodule; 174, Fusion filtering submodule; 180, Flow calculation module; 190, Result output and communication module; H, Water surface elevation; Z_b, Bed surface elevation; A(t), Cross-sectional area; B(t), Cross-sectional boundary; V_s, Water surface velocity information; Corrected cross-sectional average velocity (t) Corrected cross-sectional area Real-time traffic. Detailed Implementation

[0007] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, 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 the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description: Example 1: Real-time flow monitoring of a bank-based radar-sonar fusion system for a trapezoidal cross-section high-sediment-laden moving bed irrigation canal 1. Engineering application scenarios and overall composition This embodiment is applied to irrigation channels under conditions of high sediment content, where the channel bed is subject to rapid siltation and frequent cross-sectional changes, making traditional water level-flow rate constant curves prone to failure. The overall system structure is as follows: Figure 2 As shown, it includes: a shore-based radar measurement unit 110, a sonar measurement unit 120, a time synchronization and positioning unit 130, a data acquisition and preprocessing module 150, a dynamic profile reconstruction module 160, a fusion and correction module 170, a flow calculation module 180, and a result output and communication module 190.

[0008] The core idea of ​​this embodiment is as follows: using a shore-based radar 110 to stably and continuously acquire water surface velocity information V_s and selectable water surface elevation H; using a sonar 120 to continuously acquire water depth and bed elevation at multiple points along the cross section, and reconstructing the dynamic geometry A(t), B(t), and Z_b of the trapezoidal cross section under scouring and silting; and using a fusion and correction module 170 to adaptively update the conversion coefficient of "surface velocity → cross section average velocity", thereby realizing real-time flow calculation without the need for a fixed calibration curve.

[0009] 2. Station Layout like Figure 3 As shown, the shore-based radar measurement unit 110 is installed on the shore of one side of the channel, so that the main lobe of the radar antenna 111 covers the water surface area near the monitoring section. The radar ranging and velocity processor 112 outputs the radial velocity grid or water surface velocity characteristic value, and the water surface elevation H can be inverted from the water surface echo distance. The sonar measurement unit 120 uses a multi-point single-beam sonar measurement array 121 arranged along the cross-section direction. The sonar controller 122 polls and collects the echoes of each measurement point at a fixed period, and outputs multi-point water depth and bed elevation data. This scheme is simple to implement, cost-controllable, and has relatively more robust resistance to turbid water, making it particularly suitable for scenarios with fixed cross-sectional dimensions and limited maintenance conditions, such as irrigation channels. The time synchronization and positioning unit 130 provides a unified time reference for the radar and sonar, enabling radar observations and sonar cross-section observations to be aligned and fused within the same calculation cycle.

[0010] 3. Data Acquisition Cycle and Preprocessing In this embodiment, the following engineering sampling strategy is adopted, but it can be adjusted according to the on-site power supply and communication conditions.

[0011] The shore-based radar 110 outputs water surface velocity information V_s with an update cycle of 1 to 5 seconds; water surface elevation H can be output in intervals of 5 to 30 seconds. The sonar array 121 updates the water depth and bed elevation at multiple points along the cross section every 5 to 60 seconds. Fusion calculation cycle: Real-time traffic is output once every 10-60 seconds. With quality indicators.

[0012] The data acquisition and preprocessing module 150 performs quality control on radar and sonar data separately, including at least: Radar side: echo signal-to-noise ratio threshold elimination, velocity change identification, and water surface effective grid ratio statistics; Sonar side: echo validity judgment, abnormal water depth elimination, and missing measurement point interpolation.

[0013] After preprocessing, reliable V_s, H, and Z_b or water depth sets of multiple points on the cross section are output, providing input for subsequent cross section reconstruction and fusion correction.

[0014] 4. Dynamic Reconstruction Method for Trapezoidal Cross-Sections (Dynamic Cross-Section Reconstruction Module 160) The channel cross-section in this embodiment is trapezoidal, and its geometric elements can be characterized by: bottom width b, slope coefficient m (horizontal:vertical), and water depth h. Due to scouring and silting of the moving bed, the bed elevation Z_b changes over time, and the equivalent water depth h also changes accordingly; in addition, local scouring and silting can cause the cross-sectional shape to deviate from the ideal trapezoid. Therefore, this embodiment adopts an engineering reconstruction method of "trapezoidal prior + multi-point sonar correction": 1) Establish the trapezoidal prior section After the system is initially deployed or repaired, the trapezoidal prior parameters b_0 and m_0 are given through manual measurement or historical design parameters, and the reference range of the cross-sectional transverse coordinate x is established.

[0015] 2) Multi-point sonar cross-section point set formation The sonar measuring point array 121 outputs the bed elevation and water depth along the lateral position x_i (i=1…N), forming a cross-section point set {(x_i, z_i)}, and after removing outliers by the data acquisition and preprocessing module 150, it is input into the dynamic cross-section reconstruction module 160.

[0016] 3) Engineering solution for dynamic cross-sectional boundary B(t) and cross-sectional area A(t) When the quality of the sonar point set is good (the number of effective points ≥ the threshold N_min), the dynamic cross-section reconstruction module 160 connects the point set in a piecewise linear manner to obtain the cross-section boundary B(t), and uses discrete integration to obtain the cross-sectional area A(t) of the water passage.

[0017] When there are insufficient effective points or local missing measurements on the sonar, the dynamic profile reconstruction module 160 degenerates into "trapezoidal prior" estimation, estimating the current water depth h(t) using the effective points, and substituting the priors b_0 and m_0 into the formula. The trapezoidal area is calculated, and the "section reconstruction quality indicator" is output for the fusion and correction module 170 to adjust the weights.

[0018] 4) Output of moving bed disturbance The dynamic cross-section reconstruction module 160 calculates and outputs dynamic bed indicators such as the bed surface change ΔZ_b (compared with the mean of adjacent periods or the sliding window) and the cross-sectional area change rate dA / dt, which are used as inputs to the fusion and correction module 170.

[0019] Through the above methods, this embodiment achieves the following in engineering: priority of sonar point set and trapezoidal prior fallback, ensuring availability and continuity in the event of rapid changes in the moving bed and missing measurements.

[0020] 5. Radar-Sonar Fusion and Correction Algorithm This embodiment does not employ complex high-dimensional hydrodynamic inversion, but instead uses an engineering algorithm of "surface velocity to cross-sectional average velocity conversion coefficient + adaptive weight correction", which is simple to calculate, can be run online, and has easily tuned parameters.

[0021] 5.1 Extraction of water surface velocity features (input to velocity mapping submodule 171) The shore-based radar 110 may output multiple surface velocity grids or surface area velocity sequences. This embodiment extracts the surface velocity feature value V_s_eff(t), for example: calculating the median or truncated mean of the effective grid velocities; weighting the average by range or azimuth to obtain the surface velocity representing the vicinity of the cross section. Simultaneously, the effective grid ratio R of the radar is recorded. valid (t) is used as the quality input.

[0022] 5.2 Initial estimation of cross-sectional average velocity (velocity mapping submodule 171) In this embodiment, the water surface velocity is mapped to the initial value of the cross-sectional average flow velocity using the conversion coefficient k(t): _eff (t) The initial value k_0 of k(t) can be given by a small amount of manual flow measurement or historical experience in the early stage of system commissioning (in engineering, the initial value is usually selected between 0.75 and 0.95, which is determined by the channel roughness and flow state). It is then adaptively updated by the recursive strategy of the fusion filter submodule 174.

[0023] 5.3 Construction of Indices for Sediment-Bearing Disturbance and Moving Bed Disturbance (Moving Bed Disturbance Estimation Submodule 172, Sediment-Bearing Disturbance Estimation Submodule 173) The dynamic bed disturbance index comes from the dynamic cross-section reconstruction module 160, namely the bed surface change ΔZ_b and the cross-sectional area change rate dA / dt; the sand-containing disturbance index S_sed can be constructed in engineering by "echo intensity change", such as the normalized quantity of sonar echo intensity fluctuation and radar echo intensity fluctuation.

[0024] In this embodiment, the disturbance index is normalized to the range of 0 to 1: D bed (t): The intensity of the moving bed disturbance (normalized from |ΔZ_b| and |dA / dt|); D sed (t): Sand-bearing disturbance intensity (obtained by normalization of S_sed).

[0025] 5.4 Adaptive Correction and Fusion Output (Fusion Filtering Submodule 174) The core of this embodiment is that when the disturbance of the moving bed or sand-containing disturbance increases, the reliability of the "fixed coefficient mapping" should be reduced and a more conservative recursive update should be used instead; when the disturbance decreases and the data quality is good, k(t) is allowed to slowly adapt to maintain long-term accuracy.

[0026] Define the overall quality weight W(t): Where C sonar (t) represents the cross-section reconstruction quality, such as the effective point ratio and interpolation ratio, while w1, w2, λ1, and λ2 are the engineering tuning parameters.

[0027] 1) Cross-sectional average flow velocity correction output The corrected cross-sectional average velocity is output using a first-order exponential smoothing method: in , clip indicates clipping to ensure system stability.

[0028] 2) Output of corrected cross-sectional area of ​​water passage The cross-sectional area is preferentially adopted from the cross-sectional area A(t) output by the dynamic cross-sectional reconstruction module 160. If the sonar quality is poor, smoothing is introduced. in, Similarly, C sonar (t) and D bed (t) Adaptive change.

[0029] 3) Recursive update of transformation coefficient k(t) To avoid drastic drift in k(t) caused by short-term deviations due to moving bed and sand content, this embodiment adopts a gated recursion that updates only when the quality is good and the disturbance is low: Gating conditions: when and Updates are allowed immediately; otherwise, k(t) remains unchanged or slowly regresses to k_0, where and For manually set thresholds.

[0030] The update format is as follows: in, To integrate residual construction terms, in engineering practice, if available short-term checks are available, such as temporary velocity profiles or manual flow measurements, then... Take the difference between the measured average flow velocity and the estimated value; if there is no check, then Take the "system internal consistency residual". For example, when the cross-section is geometrically stable, i.e. when dA / dt is small, the flow rate change is required to be mainly driven by V_s. Construct residuals to keep k(t) stable.

[0031] Through the above-mentioned gating and limiting mechanism, the project can be implemented: it requires less computation, fewer parameters, and will not cause algorithm divergence due to short-term muddy water or siltation.

[0032] like Figure 4 As shown, the fusion and correction module 170 is communicatively connected to the shore-based radar measurement unit 110 and the dynamic profile reconstruction module 160, respectively. The shore-based radar measurement unit 110 provides the fusion and correction module 170 with water surface velocity information V_s and water surface elevation H, and can simultaneously provide the echo signal-to-noise ratio and the effective radar grid ratio R. valid (t) and other radar quality information; after the sonar measurement unit 120 acquires the water depth and bed elevation at multiple points along the cross section, the dynamic cross section reconstruction module 160 generates cross section geometric information (A(t), B(t)) and dynamic bed characteristics such as bed surface change ΔZ_b and cross section area change rate dA / dt, and outputs the cross section reconstruction quality index C. sonar (t) is used by the fusion and correction module 170.

[0033] The fusion and correction module 170 includes a velocity mapping submodule 171, a moving bed disturbance estimation submodule 172, a sediment-laden disturbance estimation submodule 173, and a fusion filtering submodule 174. Its workflow is as follows: The velocity mapping submodule 171 receives the surface velocity information V_s output by the shore-based radar measurement unit 110, performs robust statistics or feature extraction on the surface velocity near the cross-section to obtain a representative surface velocity V_s_eff(t), and converts V_s_eff(t) into an initial value of the cross-section average flow velocity using preset or adaptive mapping coefficients. The moving bed disturbance estimation submodule 172 receives ΔZ_b and dA / dt output by the dynamic cross-section reconstruction module (160) and generates a moving bed disturbance index to characterize the degree of rapid geometric change caused by cross-section scouring and silting. The sediment-laden disturbance estimation submodule 173 constructs a sediment-laden disturbance index based on radar echo quality characteristics and sonar echo intensity stability to characterize the degree of increased uncertainty in the echo-velocity mapping under high sediment content conditions.

[0034] The fusion filtering submodule 174 simultaneously receives the initial value of the cross-sectional average flow velocity. Cross-sectional geometric information (A(t), B(t)), radar quality information (such as R) valid (t)), cross-section reconstruction quality information (such as C) sonar (t) and disturbance indicators (such as D) bed (t), D sed (t)), and based on this, determine the adaptive fusion weights: when R valid (t) and C sonar (t) is high and D bed (t) and D sed When (t) is low, it improves the response to real-time observations and allows for slow updates to the mapping parameters; when D bed (t) or D sed (t) When the threshold is exceeded or the data quality deteriorates, the fusion weight is reduced, smoothing is enhanced, or mapping parameter updates are frozen to suppress transient anomalies caused by bed scouring and high sediment content disturbances. Finally, the fusion filtering submodule 174 outputs the corrected cross-sectional average flow velocity. With the corrected cross-sectional area The joint estimation results of (t) are provided to the traffic calculation module 180 as input for real-time traffic calculation, and the quality indicators such as fusion residual and uncertainty level are provided to the result output and communication module 190 for alarm and operation and maintenance.

[0035] 6. Flow Calculation and Output (Flow Calculation Module 180, Result Output and Communication Module 190) This embodiment calculates traffic in real time using the following simplified method: When higher accuracy is required and the radar can provide the lateral distribution of water surface velocity, a discrete integral form can also be used in the flow calculation module 180: the cross section is divided into several strips, the average velocity of the strips is obtained by V_s_eff(t) and the lateral distribution rule, and then multiplied by the local area and summed; however, this extension does not change the core structure and module division of the present invention.

[0036] The output of the results and communication module 190 should include at least the following: , , and quality indicators, such as R valid (t), C sonar W(t), fusion residual, uncertainty level, etc., and supports local display and remote transmission.

[0037] Example 2: Optimization Scheme for the Number of Sonar Arrays with Low Maintenance Deployment In scenarios where irrigation canals have narrow cross-sections or limited maintenance, the number N of sonar measurement points in the array 121 can be adjusted to 5-9, with priority given to lateral positions near the bottom and side slope transition zones to improve sensitivity to changes in the trapezoidal cross-section. The dynamic cross-section reconstruction module 160 also adopts the logic of "point set priority, prior fallback": when the point set is sufficient, piecewise linear reconstruction of B(t) is used to calculate A(t); when the point set is insufficient, trapezoidal priors b_0, m_0 and estimated water depth h(t) are used to calculate A(t).

[0038] In the fusion and correction module 170, C will be... sonar (t) is related to N: C decreases when N is small. sonar The upper limit of (t) enables the system to automatically increase the smoothing coefficient and reduce the impact of rapid cross-sectional changes on the flow rate.

[0039] Example 3: Robust Output Strategy for Periods of Heavy Scuffing and Deposition When the channel is in a period of strong scouring and silting (e.g., D) bed (t) consecutively above the threshold D for multiple periods th The system proposed in this invention can enter a "robust mode": 1) Keep k(t) frozen or slowly regress to k_0 to avoid coefficient drift; 2) Increase cross-sectional area output The smoothness intensity; 3) Output A higher uncertainty level is given, and the "strong disturbance of moving bed" status is marked in the result output and communication module 190 to prompt the operation and maintenance personnel.

[0040] This strategy does not change the system structure, but is achieved only through parameter gating of the fusion and correction module 170, the flow calculation module 180, and the result output and communication module 190, which is more in line with engineering operation and measurement reliability management.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for monitoring flow in a sand-laden, moving-bed channel using a fusion radar-sonar system, characterized in that: Includes the following steps: S101. Establish the station coordinate system and time reference: Deploy shore-based radar measurement units (110) on the bank of the channel and sonar measurement units (120) in the channel. Unify the sampling timestamps of the shore-based radar measurement units (110) and sonar measurement units (120) through the time synchronization and positioning unit (130) and establish a spatial coordinate system based on the channel cross section. S102. Obtain water surface motion, water level and water surface geometry information: The shore-based radar measurement unit (110) obtains the water surface radial velocity sequence and water surface velocity field information (V_s), and obtains the water surface echo distance information to obtain the water surface elevation (H); S103. Obtain dynamic cross-section and bed surface scouring and silting information: The sonar measurement unit (120) obtains the cross-section water depth profile and bed surface elevation sequence (Z_b) to form a dynamic cross-section geometry (A(t), B(t)) that is updated over time, where A(t) is the cross-sectional area and B(t) is the cross-sectional boundary. S104. Data preprocessing and anomaly removal: The data acquisition and preprocessing module (150) performs noise reduction, quality control, outlier removal and missing data interpolation on the shore-based radar echo and sonar echo, and outputs reliable water surface velocity information and dynamic cross-sectional information. S105. Construct a "water surface-vertical-section" velocity mapping model: In the velocity mapping submodule (171), based on the dynamic section geometry (A(t), B(t)), hydraulic elements and roughness-related parameters, establish a vertical distribution model and a lateral distribution model from the water surface velocity (V_s) to the average flow velocity of the section (V_m), and obtain the initial estimated value of the average flow velocity of the section V_m0. S106. Perform radar-sonar fusion and bed turbulence correction: In the fusion and correction module (170), a fusion correction model containing bed turbulence term is constructed using the dynamic cross-sectional change rate, bed scouring and silting index, and echo intensity characteristics. The V_m0 is adaptively corrected, and the corrected average cross-sectional velocity is output. ) and the corrected cross-sectional area ( (t)); S107. Calculate and output real-time flow rate: In the flow rate calculation module (180), calculate the flow rate based on the corrected cross-sectional average velocity ( ) and the corrected cross-sectional area ( (t) Calculate real-time traffic The system outputs real-time flow, dynamic profile and quality indicators through the result output and communication module (190).

2. The method according to claim 1, characterized in that, The shore-based radar measurement unit (110) is any one of continuous wave radar, FMCW radar, pulse Doppler radar or MIMO radar, used to output at least one of the following: water surface radial velocity, velocity spatial distribution or water surface echo distance; the sonar measurement unit (120) is any one of single beam sonar array, multi-beam sonar or phased array scanning sonar, used to acquire water depth and bed elevation at multiple measurement points along the cross section to form a dynamic cross section boundary (B(t)).

3. The method according to claim 1, characterized in that, The quality control in step S104 includes at least: elimination based on echo signal-to-noise ratio threshold, abrupt change identification based on measurement point continuity constraints, and velocity or water depth constraints based on the physically feasible range.

4. The method according to claim 1, characterized in that, The vertical distribution model in step S105 adopts a logarithmic law, power law, or stratified equivalent shear model, and introduces a correction coefficient related to sediment concentration, so that the conversion coefficient from water surface velocity to cross-sectional average flow velocity is adaptively updated according to sediment concentration and bed surface condition.

5. The method according to claim 1, characterized in that, The moving bed correction model in step S106 includes the bed surface scouring and sedimentation disturbance term ΔZ_b and the cross-sectional change rate term dA / dt, and is filtered and fused to correct the bed surface scouring and sedimentation disturbance term ΔZ_b and the cross-sectional change rate term dA / dt. and Joint estimation is performed; in step S106, a sand-containing disturbance index is constructed based on the sonar echo intensity and radar echo characteristics, and the sand-containing disturbance index S_sed is used to update the conversion coefficients in step S105 and the correction weights in step S106. The quality indicators output in step S107 include at least: flow uncertainty, cross-sectional observation coverage, radar velocity effective grid ratio, and fusion residual.

6. The method according to claim 5, characterized in that, The filtering and fusion can be any one of Kalman filtering, extended Kalman filtering, unscented Kalman filtering, or particle filtering, and the fused observations include at least water surface velocity observations, sonar profile observations, and bed surface change observations.

7. A shore-based radar-sonar fusion flow monitoring system for a channel with a sand-laden moving bed, characterized in that, include: The system includes: a shore-based radar measurement unit (110) for acquiring water surface velocity and water level information; a sonar measurement unit (120) for acquiring dynamic cross-sectional water depth and bed elevation information; a time synchronization and positioning unit (130) for providing a unified time and space reference for the shore-based radar measurement unit (110) and the sonar measurement unit (120); a data acquisition and preprocessing module (150) for denoising, quality control, and anomaly removal of the original radar and sonar data; a dynamic cross-sectional reconstruction module (160) for reconstructing and updating the cross-sectional area and boundary based on sonar observations; a fusion and correction module (170) for adaptively correcting the mapping from water surface velocity to average cross-sectional velocity based on dynamic bed disturbance and sediment-laden disturbance; a flow calculation module (180) for calculating real-time flow based on the corrected velocity and cross-section; and a result output and communication module (190) for outputting flow, cross-sectional and quality indicators and displaying them locally or transmitting them remotely. Among them, the fusion and correction module (170) and the dynamic cross-section reconstruction module (160) work together to enable the system to output real-time flow even under the condition of continuous scouring and silting of the cross-section.

8. The system according to claim 7, characterized in that... The sonar measurement unit (120) includes a sonar measurement point array (121) arranged along the cross section and a sonar controller (122). The sonar measurement point array (121) is fixedly installed on the cross-channel support component and submerged in the water. The sonar controller (122) is installed in a protective box on the shore. The sonar measurement point array (121) is used to provide water depth and bed elevation at multiple points along the cross section. The sonar controller (122) is used for scanning control and echo calculation.

9. The system according to claim 7, characterized in that, The fusion and correction module (170) includes: a velocity mapping submodule (171), a moving bed disturbance estimation submodule (172), a sand-containing disturbance estimation submodule (173), and a fusion filtering submodule (174), used to correct the average flow velocity of the cross-section ( ) and the corrected cross-sectional area ( The joint estimation results of (t)).

10. The system according to any one of claims 7 to 9, characterized in that, The fusion and correction module (170) is signal-connected to the shore-based radar measurement unit (110) and the dynamic cross-section reconstruction module (160) to acquire water surface velocity information (V_s), cross-section geometric information (A(t), B(t)), and bed elevation information (Z_b). The fusion and correction module (170) includes: a velocity mapping submodule (171) for generating an initial cross-section average flow velocity estimate (V_m0) based on the water surface velocity information (V_s) output by the shore-based radar measurement unit (110); and a moving bed disturbance estimation submodule (172) for estimating the dynamic cross-section based on the water surface velocity information (V_s) output by the dynamic cross-section reconstruction module. (160) The cross-sectional area change rate (dA / dt) or bed surface change (ΔZ_b) output is used to generate the moving bed disturbance index; the sediment-laden disturbance estimation submodule (173) is used to generate the sediment-laden disturbance index based on the echo quality characteristics of the shore-based radar measurement unit (110) and the echo quality characteristics of the sonar measurement unit (120); the fusion filtering submodule (174) is used to adaptively fuse the cross-sectional average velocity and cross-sectional area based on the initial cross-sectional average velocity estimate (V_m0), cross-sectional geometric information (A(t), B(t)) and the moving bed disturbance index and the sediment-laden disturbance index.

Images