Satellite big data-driven dynamic measurement and calculation method for section area of watercourse without hydrologic station

Abstract

The invention relates to the technical field of water conservancy information, in particular to a satellite big data driven non-hydrometric station river section area dynamic measurement method, which comprises the following steps: calculating a river width and radar phase path deduced gradient, constructing a roughness inversion system coupling section geometry and hydraulic balance, traversing river bottom elevation parameters, and calculating the river bottom elevation. According to the method, by adopting multi-spectral image sub-pixel decomposition and radar phase path optimization, inversion of river surface width and water surface gradient data is carried out, geographical environment limitation is broken through, monitoring timeliness is improved, resistance inversion logic based on hydraulic balance is constructed, and the real river bottom elevation is locked according to the principle of roughness time sequence statistics minimum dispersion. And global search and automatic locking are performed on river bottom elevation parameters by using statistical stability characteristics of a river course roughness coefficient on a time sequence, so that dynamic inversion is performed on a river course water passing area according to remote sensing data, and data support is provided for drainage basin water resource management.

Description

Technical Field

[0001] This invention relates to the field of water conservancy information technology, and in particular to a method for dynamically calculating the cross-sectional area of ​​river channels without hydrological stations, driven by satellite big data. Background Technology

[0002] The field of water conservancy information technology involves various system engineering projects that utilize modern sensor communication and computer technologies to digitally collect, transmit, process, and apply hydrological and water resources data. Its core objective is to construct a three-dimensional sensing network covering the watershed scale by integrating remote sensing monitoring, geographic information systems, and hydrological telemetry terminals. This enables real-time monitoring and simulation analysis of key elements such as rainfall, water level, flow, and water quality, providing data support and decision-making assistance for water resource allocation, flood control and disaster reduction, and water ecological protection. Among these methods, the traditional dynamic measurement method for river cross-sectional area without hydrological stations refers to the technical task of geometrically measuring and estimating the flow area of ​​river cross-sections lacking fixed ground monitoring stations. This typically involves manually carrying a total station or RTK equipment to the site to measure the riverbank topography and using a sounding rod or echo sounder to obtain underwater topographic data to generate a measured large-scale river cross-section map. The flow area is then calculated using a geometric cutting method based on water level data, or by using the Manning formula combined with river gradient and roughness parameters to estimate the flow area at a specific water level.

[0003] Traditional field surveying relies on personnel carrying heavy equipment to conduct on-site operations, resulting in high time costs and safety risks in remote or complex river sections. This limits the spatial coverage and update frequency of monitoring data. Relying on static geometric cutting methods often lacks timely measured underwater topographic data. Hydraulic calculations based on empirical formulas are difficult to obtain accurate roughness parameters in areas without measured data. This leads to significant deviations in the calculation results of the water flow area when the river channel boundary changes or the water level fluctuates greatly. This makes it difficult to meet the actual needs of high-frequency dynamic monitoring of water resources in large-scale watersheds, resulting in spatiotemporal blind spots and accuracy bottlenecks in the acquisition of hydrological information. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a method for dynamically calculating the cross-sectional area of ​​river channels without hydrological stations, driven by satellite big data, comprising the following steps:

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a dynamic calculation method for river cross-sectional area without hydrological stations driven by satellite big data, comprising the following steps:

[0006] S1: Analyze the multispectral satellite image data of the river section to be monitored and construct the water-land boundary skeleton. Extract the pure water reference spectrum and pure soil reference spectrum along the normal direction of the water-land boundary skeleton, construct a linear hybrid decomposition model, calculate the proportion of sub-pixel level water body, and obtain the water surface width data of the river section.

[0007] S2: Acquire radar interferometric phase data and river centerline vector data, construct a streamline coordinate system, map the radar interferometric phase data to the streamline coordinate system, set the reverse flow impedance cost to be greater than the forward flow impedance cost, search for the phase path with the minimum cumulative impedance cost, calculate the phase change rate and acquire water surface gradient data.

[0008] S3: Construct a river channel geometric relationship model, set the search space for riverbed elevation parameters and generate multiple riverbed elevation trial data, combine the water surface width data with the river channel geometric relationship model, calculate the cross-sectional area sequence and hydraulic radius sequence, and obtain the river channel cross-sectional parameter set;

[0009] S4: Based on the physical balance mechanism between the gravity component of the water flow and the frictional resistance of the riverbed boundary, a fluid resistance inversion model is constructed. The water surface gradient data and the set of geometric parameters of the river cross section are substituted into the fluid resistance inversion logic for reverse deduction. The riverbed resistance coefficient sequence is calculated, and the roughness coefficient data is obtained.

[0010] S5: Calculate the statistical dispersion of the roughness coefficient data, select the riverbed elevation trial data corresponding to the smallest dispersion term and mark it as the optimal riverbed elevation data, substitute it into the river channel geometric relationship model, calculate the cross-sectional area data, and obtain the area data measurement record.

[0011] As a further aspect of the present invention, the water surface width data includes sub-pixel level river width values ​​and sub-pixel coordinates of the water-land interface; the water surface gradient data includes water surface slope values ​​along the course and flow direction phase gradient components; the river channel cross-section parameter set includes exploratory riverbed elevation samples, multiple hypothetical water-passing area matrices, and corresponding hydraulic radius matrices; the roughness coefficient data includes inverted Manning roughness sequence and riverbed comprehensive resistance values; and the area data measurement record includes target time water-passing area values, locked riverbed elevation parameters, and basic information for flow monitoring.

[0012] As a further aspect of the present invention, the step of obtaining the water surface width data specifically includes:

[0013] S101: Acquire multispectral satellite image data of the river section to be monitored, perform band channel spatial convolution processing on the multispectral satellite image data, calculate pixel spatial gradient distribution features and construct image spectral gradient distribution layer, filter local maximum gradient modulus data in the image spectral gradient distribution layer, connect pixel points according to the river flow direction connectivity rules, and construct high gradient boundary skeleton data.

[0014] S102: Calculate the gradient vector normal direction of the edge pixels in the high gradient boundary skeleton data, search for homogeneous pixels with the minimum gradient modulus in the neighborhood along the normal direction, screen the pure background pixels with the smallest Euclidean distance between the water side and the land side, extract the spectral response features of the pure background pixels, and establish endmember spectral reference data containing pure water spectral features and pure soil spectral features.

[0015] S103: Call the end-member spectral reference data, analyze the composition of mixed pixel components and construct a linear spectral unmixing model, substitute the multispectral satellite image data into the linear spectral unmixing model to perform sub-pixel decomposition calculation, calculate the water component proportion data in a single pixel, combine the spatial resolution parameters to convert the water component proportion data into entity area and perform horizontal accumulation to obtain the water surface width data of the river cross section.

[0016] As a further aspect of the present invention, the step of obtaining the water surface gradient data specifically includes:

[0017] S201: Obtain radar interferometric phase data and river centerline vector data of the river section to be monitored. Construct a streamline coordinate system with the river centerline as the vertical axis. Calculate the projection coordinates and normal deviation data of the pixels in the radar interferometric phase data relative to the vertical axis. Based on the projection coordinates, resample and map the original phase data to the grid nodes of the streamline coordinate system to establish the streamline system phase mapping matrix.

[0018] S202: Call the streamline system phase mapping matrix, set the impedance cost of the phase jump in the counter-current direction to be greater than the impedance cost of the phase jump in the downstream direction, search for the pixel connection path with the minimum accumulated impedance cost in the streamline system phase mapping matrix, connect adjacent pixel nodes according to the pixel connection path and smooth the local phase discontinuity region to obtain the flow direction continuous phase sequence.

[0019] S203: Perform differential calculation along the flow direction for the continuous phase sequence, obtain phase gradient data between adjacent nodes and construct the phase change rate curve along the flow direction, calculate the phase elevation conversion coefficient by combining the radar carrier wavelength parameter and the observation incident angle parameter, and obtain the water surface gradient data of the river section.

[0020] As a further aspect of the present invention, the steps for obtaining the river channel cross-section parameter set are specifically as follows:

[0021] S301: Based on the relationship between the river surface width and the effective water depth, construct a geometric relationship model of the river channel, analyze the range of geographical elevation variation of the river section to be monitored, and set the lower limit and upper limit of the search for the riverbed elevation parameter. Discretize the sampling at fixed intervals between the lower limit and the upper limit of the search to generate a sequence of riverbed elevation trial samples.

[0022] S302: Call the water surface width data, take each elevation assumption item in the riverbed elevation test sample sequence as the riverbed reference plane, use the river channel geometric relationship model to reverse deduce the effective water depth data for each water surface width, perform geometric integration on the effective water depth data, calculate the water flow area under various elevation assumption conditions, and establish the water flow cross-sectional area matrix.

[0023] S303: Calculate the wetted perimeter length of each area data in the cross-sectional area matrix corresponding to the water passage, calculate the hydraulic radius data corresponding to each elevation assumption, and associate and combine the multi-assumption cross-sectional area matrix with the hydraulic radius data to obtain the river channel cross-sectional parameter set.

[0024] As a further aspect of the present invention, the process of constructing a river channel geometric relationship model based on the relationship between river surface width and effective water depth specifically includes:

[0025] Based on the hydraulic geometry theory of natural river channel cross-sections, a power function form is selected as the basic mathematical structure for the lateral expansion characteristics of the river channel, and a cross-sectional morphology equation is established with effective water depth as an independent variable and water surface width as a non-independent variable.

[0026] Geometric integration is performed on the cross-sectional shape equation to derive the formula for calculating the cross-sectional area of ​​the water passage corresponding to the target geometry. Based on the principle of curve arc length integration, a formula for calculating the wetted perimeter is constructed, and an algebraic relationship logic between the cross-sectional area, wetted perimeter, and hydraulic radius is established.

[0027] The riverbed elevation parameter to be optimized is defined as the vertical starting datum of the cross-sectional geometric profile. The relative positional relationship between the effective water depth and the riverbed elevation parameter is set in the algebraic association logic to construct a full cross-sectional geometric description system containing vertical datum information.

[0028] For complex riverbed landforms, a piecewise function is used to construct geometric mapping rules for the main channel area and the floodplain area respectively. By splicing the morphological equations of different water level intervals, a parametric channel geometric relationship model that continuously describes the evolution of cross-sectional shape from the dry season to the wet season is generated.

[0029] As a further aspect of the present invention, the step of obtaining the roughness coefficient data specifically includes:

[0030] S401: Based on the physical balance mechanism between the gravity component of the water flow and the frictional resistance of the riverbed boundary, the motion law of the open channel fluid under gravity drive is analyzed, the dynamic equations relating the hydraulic radius, cross-sectional area and energy gradient are constructed, the resistance coefficient is set as the dependent variable to be solved and the calculation rules for inferring the resistance characteristics using geometric parameters are established, and a fluid resistance inversion model is constructed.

[0031] S402: Call the river cross-section parameter set, extract the cross-sectional area sequence and hydraulic radius sequence corresponding to different riverbed elevation assumptions, and perform data association and dimension alignment between the water surface gradient data and the cross-sectional area sequence and hydraulic radius sequence to generate the hydrodynamic calculation input sequence;

[0032] S403: Substitute the hydrodynamic calculation input sequence into the fluid resistance inversion model for inverse calculation, calculate the theoretical resistance parameters required to maintain the current water flow state for each riverbed elevation trial data, traverse the elevation parameter search space and generate a resistance coefficient distribution sequence corresponding to various riverbed elevation assumptions, and obtain roughness coefficient data.

[0033] As a further aspect of the present invention, the process of analyzing the motion law of open channel fluid under gravity driven by gravity based on the physical balance mechanism between the gravity component of water flow along the flow path and the frictional resistance of the riverbed boundary, constructing a dynamic equation relating the hydraulic radius, cross-sectional area, and energy gradient, setting the resistance coefficient as the dependent variable to be solved, and establishing calculation rules for inferring resistance characteristics using geometric parameters to construct a fluid resistance inversion model is specifically as follows:

[0034] The force equilibrium state of the water flow in the monitored river section under the action of gravity is analyzed. The numerical balance relationship between the gravity component along the direction of water flow and the frictional resistance generated by the riverbed boundary is set and used as a physical constraint condition for fluid motion.

[0035] Based on the physical constraints, a hydraulic control equation describing the uniform flow motion state of an open channel is constructed. A function mapping logic is set that the average flow velocity at the cross section is inversely proportional to the riverbed resistance coefficient and directly proportional to the power function factor of the hydraulic radius and the square root term of the energy gradient.

[0036] Based on the mass conservation condition of water flow, a continuity equation is constructed in which the instantaneous flow rate of the cross-section is equal to the product of the cross-sectional area and the average flow velocity of the cross-section. The continuity equation is then used to substitute variables in the hydraulic control equation to eliminate the flow velocity variable in the equation.

[0037] The equations after variable substitution are analyzed in reverse, and the riverbed resistance coefficient is separated into independent terms to be solved. A fluid resistance inversion model is established with the cross-sectional area, the power function factor of the hydraulic radius, the square root term of the energy gradient, and the instantaneous flow rate of the cross-section as calculation factors.

[0038] As a further aspect of the present invention, the steps for obtaining the area data measurement record are specifically as follows:

[0039] S501: Perform time-dimensional stability analysis on the roughness coefficient data, calculate the fluctuation amplitude of the resistance coefficient sequence corresponding to various riverbed elevation assumptions in multi-temporal observations, quantitatively evaluate the discrete characteristics of each sequence, and construct time series statistical dispersion data.

[0040] S502: Analyze the time series statistical dispersion data and perform a minimum value search, select the riverbed elevation trial sample corresponding to the item with the smallest dispersion, and determine it as the optimal riverbed elevation data;

[0041] S503: Substitute the optimal riverbed elevation data into the river channel geometric relationship model to lock the model parameters, substitute the water surface width data at the current monitoring time into the locked model for calculation, calculate the cross-sectional area data at the current time, and obtain the area data measurement record.

[0042] As a further aspect of the present invention, the process of substituting the optimal riverbed elevation data into the river channel geometric relationship model to lock the model parameters, substituting the water surface width data at the current monitoring time into the locked model for calculation, and calculating the cross-sectional area data of the water flow at the current time is specifically as follows:

[0043] The optimal riverbed elevation data is set as the vertical starting point of the channel geometric relationship model. It is substituted into the channel geometric relationship model to determine the zero point of the effective water depth variable, and the numerical locking of the undetermined geometric parameters in the model is performed to generate a unique and definite cross-sectional shape equation for the river section to be monitored.

[0044] The water surface width data at the current monitoring time is called up, and based on the function characteristic of the water surface width changing monotonically with the effective water depth in the cross-sectional shape equation, the real-time effective water depth value corresponding to the current water surface width value is calculated.

[0045] An integral operation rule with effective water depth as the independent variable is set, the lower limit of the integration interval is set to zero water depth, and the upper limit is set to the real-time effective water depth value. A definite integral calculation is performed on the cross-sectional shape equation to obtain the geometric area of ​​the cross-section under the current water level conditions, and the cross-sectional area data is output.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0047] In this invention, by employing multispectral image sub-pixel decomposition and radar phase path optimization, the river width and water surface gradient data are inverted, overcoming geographical limitations and improving monitoring timeliness. A resistance inversion logic based on hydraulic balance is constructed, and by utilizing the statistical stability characteristics of the river roughness coefficient over time, the riverbed elevation parameters are globally searched and automatically locked, enabling dynamic inversion of the river channel's water-passing area based on remote sensing data, thus providing data support for watershed water resources management. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the steps of the present invention;

[0050] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0051] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0052] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0053] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0054] Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0055] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0056] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0057] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0058] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0059] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0060] Please see Figure 1 This invention provides a method for dynamically calculating the cross-sectional area of ​​a river channel without hydrological stations, driven by satellite big data, including the following steps:

[0061] S1: Analyze the multispectral satellite image data of the river section to be monitored and construct the water-land boundary skeleton. Extract the pure water reference spectrum and pure soil reference spectrum along the normal direction of the water-land boundary skeleton, construct a linear hybrid decomposition model, calculate the proportion of sub-pixel level water body, and obtain the water surface width data of the river section.

[0062] S2: Acquire radar interferometric phase data and river centerline vector data, construct a streamline coordinate system, map the radar interferometric phase data to the streamline coordinate system, set the reverse flow impedance cost to be greater than the forward flow impedance cost, search for the phase path with the minimum cumulative impedance cost, calculate the phase change rate and acquire water surface gradient data.

[0063] S3: Construct a river channel geometric relationship model, set the search space for riverbed elevation parameters and generate multiple riverbed elevation trial data, combine the water surface width data with the river channel geometric relationship model, calculate the cross-sectional area sequence and hydraulic radius sequence, and obtain the river channel cross-sectional parameter set;

[0064] S4: Based on the physical balance mechanism between the gravity component of water flow and the frictional resistance of the riverbed boundary, a fluid resistance inversion model is constructed. The water surface gradient data and the set of geometric parameters of the river cross section are substituted into the fluid resistance inversion logic to reverse the deduction, calculate the riverbed resistance coefficient sequence, and obtain the roughness coefficient data.

[0065] S5: Calculate the statistical dispersion of the roughness coefficient data, select the riverbed elevation trial data corresponding to the smallest dispersion term and mark it as the optimal riverbed elevation data, substitute it into the river channel geometric relationship model, calculate the cross-sectional area data, and obtain the area data measurement record.

[0066] The water surface width data includes sub-pixel level river width values ​​and sub-pixel coordinates of the water-land interface; the water surface gradient data includes water surface slope values ​​along the course and flow direction phase gradient components; the river channel cross-section parameter set includes exploratory riverbed elevation samples, multi-hypothetical water-passing area matrices, and corresponding hydraulic radius matrices; the roughness coefficient data includes the inverted Manning roughness sequence and riverbed comprehensive resistance values; and the area data calculation records include the target time water-passing area setpoint, the locked version of the riverbed elevation parameters, and basic information on flow monitoring.

[0067] Please see Figure 2 The specific steps for obtaining water surface width data are as follows:

[0068] S101: Acquire multispectral satellite image data of the river section to be monitored, perform band channel spatial convolution processing on the multispectral satellite image data, calculate pixel spatial gradient distribution features and construct image spectral gradient distribution layer, filter local maximum gradient modulus data in the image spectral gradient distribution layer, connect pixel points according to the river flow direction connectivity rules, and construct high gradient boundary skeleton data.

[0069] For the river section to be monitored, Sentinel-2 multispectral satellite imagery data of the corresponding time phase was selected as the processing object. The near-infrared band with a spatial resolution of 10 meters was retrieved as the main analysis channel, and the two-dimensional pixel matrix of this band was applied with a size of [missing information]. The Sobel convolution operator is used for full-frame scanning, calculating the grayscale change rate of each pixel in the horizontal and vertical directions. The gradient values ​​in the two directions are squared and squared to obtain the gradient modulus value of each pixel position, generating an image spectral gradient distribution layer with the same row and column dimensions. The gradient modulus threshold is set to the 85th percentile of the gradient values ​​of all pixels in the image, for example, the gradient modulus threshold is set to 45. The layer is traversed to filter out local maxima pixels with gradient modulus greater than 45. These high gradient pixels are defined as candidate boundary points. According to the continuity constraint of water flow direction, the eight-neighborhood connectivity algorithm is used to track and connect the candidate boundary points, and isolated noise line segments with a length of less than 50 pixels are removed. The set of the remaining continuous pixel connections is established as the high gradient boundary skeleton data. This skeleton data accurately depicts the boundary contour of water and land in the image space.

[0070] S102: Calculate the gradient vector normal direction of edge pixels in the high gradient boundary skeleton data, search for homogeneous pixels with the minimum gradient modulus in the neighborhood along the normal direction, screen the pure background pixels with the smallest Euclidean distance between the water side and the land side, extract the spectral response features of the pure background pixels, and establish endmember spectral benchmark data containing pure water spectral features and pure soil spectral features.

[0071] For the constructed high-gradient boundary skeleton data, each edge pixel on the skeleton is traversed one by one. The perpendicular direction of the pixel along the boundary tangent, i.e., the gradient vector normal direction, is determined using the difference calculation method. A linear window with a search radius of 5 pixels is set in this normal direction, extending towards both the water and land sides. The variance of the spectral reflectance of each pixel within the window is calculated. Pixel sets with a variance less than 0.005 are classified as homogeneous regions. Within the homogeneous region on the water side, the pixel with the closest Euclidean distance to the skeleton pixel and the normalized water index N is selected. Pixels with a DWI greater than 0.3 are marked as pure water reference pixels. Within the homogeneous region on the land side, pixels with the closest Euclidean distance to the skeleton pixels and a Normalized Difference Vegetation Index (NDVI) greater than 0.2 are selected and marked as pure soil reference pixels. The reflectance values ​​of the pure water reference pixels in the near-infrared band are extracted, for example, 0.05, and the reflectance values ​​of the pure soil reference pixels in the near-infrared band are extracted, for example, 0.25. These two values ​​are paired to form the endmember spectral reference data for the edge point, as shown in Table 1, thus completing the endmember feature calibration required for the decomposition of mixed pixels.

[0072] S103: Call the end-member spectral reference data, analyze the composition of mixed pixel components and construct a linear spectral unmixing model, substitute the multispectral satellite image data into the linear spectral unmixing model to perform sub-pixel decomposition calculation, calculate the water component proportion data in a single pixel, combine the spatial resolution parameters to convert the water component proportion data into the entity area and perform horizontal accumulation to obtain the water surface width data of the river cross section.

[0073] Using endmember spectral reference data, a mathematical analytical expression based on linear spectral mixing theory is established for each mixed pixel on the high-gradient boundary skeleton. The spectral observation value of the mixed pixel is set as follows: The reflectivity of pure water end-members is The reflectivity of pure earth end-members is Establish equations ,in To determine the proportion of water components, the actual observed reflectance of the mixed pixel, for example, 0.15, is substituted into the equation. The calculated water area proportion within that pixel is 0.5, or 50%. This is then combined with the spatial resolution parameters of the satellite imagery, such as the actual ground area corresponding to a single pixel. The area of ​​water within the mixed pixel is calculated to be 50 square meters. The area of ​​all pure water pixels within the same scan row of the river cross-section is horizontally summed with the area of ​​the mixed pixel water area after decomposition. If a cross-section row contains 20 pure water pixels and 2 edge-mixed pixels, the summation result is... The area is calculated in square meters. The total area is divided by the spatial resolution step of pixels, which is 10 meters. The sub-pixel level precision water surface width data at this cross-section is 210 meters, thus completing the accurate measurement of the water surface width.

[0074] Table 1. Sampling Table of Endmember Spectral Reference Data

[0075]

[0076] See Table 1, which lists the endmember reflectance and mixed pixel observations extracted at different skeleton locations to support linear spectral unmixing calculations. This data indicates that the background spectrum at different locations is heterogeneous and needs to be extracted locally and dynamically.

[0077] Please see Figure 3 The specific steps for obtaining water surface gradient data are as follows:

[0078] S201: Obtain radar interferometric phase data and river centerline vector data of the river section to be monitored. Construct a streamline coordinate system with the river centerline as the vertical axis. Calculate the projection coordinates and normal deviation data of the pixels in the radar interferometric phase data relative to the vertical axis. Based on the projection coordinates, resample and map the original phase data to the grid nodes of the streamline coordinate system to establish the streamline system phase mapping matrix.

[0079] The Sentinel-1 radar interferometric phase data for the river section to be monitored is loaded. This data includes entangled phase information after removing the flat-land effect. Simultaneously, the river centerline vector data extracted through a geographic information system is loaded, with the upstream endpoint of the river centerline as the origin and the tangential direction along the centerline defined as the vertical axis. The direction perpendicular to the centerline is defined as the horizontal axis. Construct an orthogonal streamline coordinate system, traverse every pixel in the radar phase image, and read its geographic coordinates. Calculate the projected position of this coordinate point in the streamlined coordinate system. Calculate the normal distance deviation between the projected point and the original pixel. The maximum effective deviation threshold is set to 1.5 times the river width, for example, 300 meters. Noise pixels with a deviation greater than 300 meters are removed. A bilinear interpolation algorithm is used, based on the streamline coordinate system grid nodes. The position is determined by resampling the corresponding phase values ​​in the original phase image to generate a streamlined phase mapping matrix with a resolution of 10m×10m, thereby achieving geometric correction of the phase data from the Cartesian coordinate system to the hydraulic streamlined coordinate system.

[0080] S202: Call the streamline system phase mapping matrix, set the impedance cost of the phase jump in the counter-current direction to be greater than the impedance cost of the phase jump in the downstream direction, search for the pixel connection path with the minimum cumulative impedance cost in the streamline system phase mapping matrix, connect adjacent pixel nodes according to the pixel connection path and smooth the local phase discontinuity region to obtain the flow direction continuous phase sequence.

[0081] In the streamline system phase mapping matrix, the phase jump impedance function between adjacent pixels is defined, and the downstream direction (i.e., The expected phase gradient in the positive direction (on the axis) is negative, and in the reverse direction it is positive, for two adjacent nodes. and If from point to In the downstream direction and phase difference exist Within the range, set the impedance cost. ,like For positive values ​​or values ​​outside the range, a penalty coefficient is introduced. Setting impedance cost If from point to For the counter-current direction, the impedance cost is directly set to 5 times that for the downstream direction. Dijkstra's shortest path search algorithm is applied to find a pixel connection path in the matrix that minimizes the cumulative impedance cost from the upstream boundary to the downstream boundary. This path is the phase-continuous path least affected by speckle noise. All pixel nodes are connected along this path. For phase jumps exceeding [a certain threshold] on the path... For the local discontinuous regions, a mean filter with a window size of 5 is used for smoothing correction to obtain a smooth and continuous flow-direction continuous phase sequence.

[0082] S203: Perform differential calculation along the flow direction for the continuous phase sequence, obtain phase gradient data between adjacent nodes and construct the phase change rate curve along the flow, and calculate the phase elevation conversion coefficient by combining the radar carrier wavelength parameters and the observation incident angle parameters to obtain the water surface gradient data of the river section.

[0083] Perform along-path numerical difference operations on the acquired continuous phase sequence of the flow direction to calculate the phase difference value between adjacent nodes. Difference between streamline distance and streamline ,ratio This involves using phase gradient data along the entire river length to perform linear regression fitting and extract the slope of the phase change rate. For example, the calculated phase change rate Radius / meter, retrieve the system parameters of the radar sensor, including the radar carrier wavelength. meters, radar beam incident angle The phase elevation conversion coefficient is calculated using the interferometric elevation conversion formula. Substituting the numerical values, we obtain meters per radian, the rate of phase change Multiply by conversion factor The calculated rate of change of water surface elevation along the path is... That is, the water surface gradient data for this section of the river is This value reflects the intensity of the driving force of the water flow.

[0084] Please see Figure 4 The specific steps for obtaining the river channel cross-section parameter set are as follows:

[0085] S301: Based on the relationship between the river surface width and the effective water depth, construct a geometric relationship model of the river channel, analyze the range of geographical elevation variation of the river section to be monitored, and set the lower limit and upper limit of the search for the riverbed elevation parameter. Discretize the sampling at fixed intervals between the lower limit and the upper limit of the search to generate a sequence of riverbed elevation trial samples.

[0086] Based on the geometric generalization theory of trapezoidal or compound channels in hydraulics, a method for describing river width is constructed. With effective water depth A changing river channel geometry model is used to retrieve DEM (Digital Elevation Model) data for the area to be monitored. The topographic elevation distribution on both sides of the riverbank is analyzed to determine the potential distribution range of the riverbed elevation. For example, if the DEM shows a riverbank elevation of 20 meters, and historical hydrological records suggest a maximum water depth of no more than 10 meters, then the lower limit for searching the riverbed elevation is set to 10 meters, and the upper limit to 20 meters. Within this range, a sampling interval of 0.1 meters is set to generate a sequence of 100 discrete riverbed elevation trial samples. This sequence covers all possible hypotheses regarding the vertical location of the riverbed.

[0087] S302: Call the water surface width data, take each elevation assumption in the riverbed elevation test sample sequence as the riverbed reference plane, use the river channel geometric relationship model to reverse deduce the effective water depth data for each water surface width, perform geometric integration on the effective water depth data, calculate the water flow area under various elevation assumption conditions, and establish the water flow cross-sectional area matrix.

[0088] Retrieve water surface width data, for example, the water surface width measured at a certain moment. meters, the water level at that time was (This can be obtained through satellite altimetry, for example, 25 meters), for each hypothetical value in the sequence of riverbed elevation test samples. ,For example Meters, calculate the corresponding effective water depth If a power function model is used, the meter... To describe the cross-sectional morphology, parameters need to be derived by combining historical width-water level data from multiple periods. and Here, we directly utilize the definition of integral, assuming the river channel is a regular parabola, and the area of ​​the cross-section of the water passage... It can be obtained by integrating the width function, i.e. Under discrete conditions, it simplifies to Substituting the numerical values, we get The calculation process is repeated for each riverbed elevation sample, and the above calculation process is repeated to generate a cross-sectional area matrix containing 100 elements, where each element corresponds to a specific riverbed elevation assumption.

[0089] S303: Calculate the wetted perimeter of each area data in the corresponding cross-sectional area matrix, calculate the hydraulic radius data corresponding to each elevation assumption, and combine the multi-assumption cross-sectional area matrix with the hydraulic radius data to obtain the river cross-sectional parameter set.

[0090] For each cross-sectional geometry under the given riverbed elevation assumption, calculate its corresponding wetted perimeter. Based on the approximate formula for the wetted perimeter of a parabolic channel Substitute Mihe meters, calculated Meters, using the hydraulic radius definition formula The area calculated above Substitute the values ​​and calculate the hydraulic radius. meters, the calculated area sequence With hydraulic radius sequence By matching and combining the indexes of the riverbed elevation assumptions one-to-one, a set of river cross-section parameters containing multiple geometric assumptions is constructed, providing a complete geometric input library for subsequent hydraulic inversion.

[0091] Please see Figure 5 The specific steps for obtaining roughness coefficient data are as follows:

[0092] S401: Based on the physical balance mechanism between the gravity component of the water flow and the frictional resistance of the riverbed boundary, the motion law of the open channel fluid under gravity drive is analyzed, the dynamic equations relating the hydraulic radius, cross-sectional area and energy gradient are constructed, the resistance coefficient is set as the dependent variable to be solved and the calculation rules for inferring the resistance characteristics using geometric parameters are established, and a fluid resistance inversion model is constructed.

[0093] Based on the simplified form of Saint-Venant's equations for steady non-uniform flow in open channels, i.e., a variant of Manning's formula, this paper analyzes the flow under the influence of gravity components. Boundary friction The equilibrium relationship between them is used to construct the dynamic equations. ,in For traffic, For roughness, For area, For hydraulic radius, To reduce the gradient, a mathematical transformation is performed on the equation, changing the riverbed roughness coefficient. Define the dependent variable as the left-hand side of the equation and establish the operational rules. This rule utilizes geometric parameters ( ) and hydraulic parameters ( To infer the physical roughness of the riverbed, a fluid resistance inversion model is constructed. Here, daily average flow estimates from global hydrological reanalysis datasets (such as GloFAS) are used as model inputs. ,For example cubic meters per second, thus closing the solution path for the equation.

[0094] S402: Call the river cross-section parameter set, extract the cross-sectional area sequence and hydraulic radius sequence corresponding to different riverbed elevation assumptions, associate the water surface gradient data with the cross-sectional area sequence and hydraulic radius sequence, and generate the hydrodynamic calculation input sequence.

[0095] Call the river cross-section parameter set and extract the parameters corresponding to the assumed riverbed elevation. A set of data in meters, namely the cross-sectional area of ​​the water passage. square meters, hydraulic radius Meters, while simultaneously calling the water surface gradient data obtained from S203. These heterogeneous data sources are aligned in terms of both time and spatial dimensions to ensure that all parameters describe the water flow state of the same river segment at the same time. For time series data, linear interpolation is used to interpolate the flow data. Align with satellite transit times and construct a system containing The hydrodynamic calculation input sequence consists of four sets of vectors, with each row of data corresponding to a specific riverbed elevation assumption.

[0096] S403: Substitute the hydrodynamic calculation input sequence into the fluid resistance inversion model for inverse calculation, calculate the theoretical resistance parameters required to maintain the current water flow state for each riverbed elevation trial data, traverse the elevation parameter search space and generate the resistance coefficient distribution sequence corresponding to various riverbed elevation assumptions, and obtain roughness coefficient data.

[0097] The constructed hydrodynamic calculation input sequence is substituted into the constructed fluid resistance inversion model, taking into account the riverbed elevation assumption. In the case of meters, substitute the values ​​for calculation, first calculate the hydraulic radius term.

[0098] Calculate the reduction term Substitute into the inversion formula The theoretical roughness coefficient under this assumption is calculated. Iterate through all 100 riverbed elevation trial data generated by S301, repeating the above calculation steps to obtain the resistance coefficient distribution sequence corresponding to different riverbed elevation assumptions. For example, when assuming the riverbed elevation is near the true value, the calculated resistance coefficient distribution sequence is... The value should be within a reasonable physical range of 0.025 to 0.045, and assumptions that deviate from the true value will lead to The values ​​exhibit physically unreasonable extreme values, thus obtaining a set of roughness coefficient data corresponding to multiple hypotheses.

[0099] Please see Figure 6 The specific steps for obtaining area measurement records are as follows:

[0100] S501: Perform time-dimensional stability analysis on roughness coefficient data, calculate the fluctuation amplitude of the resistance coefficient sequence corresponding to various riverbed elevation assumptions in multi-temporal observations, quantitatively evaluate the discrete characteristics of each sequence, and construct time series statistical dispersion data.

[0101] A long-term statistical analysis was performed on the roughness coefficient data, selecting 24 satellite observation data periods acquired within the past 12 months, for each riverbed elevation assumption (e.g., (meters), extracting the roughness coefficient values ​​calculated at 24 different time phases under this assumption, and constructing a time series vector. Calculate the arithmetic mean of the vector. and standard deviation Using the formula Calculate the coefficient of variation to quantify and evaluate the discrete characteristics of the sequence, for example, for... meters, calculated ,for meters, calculated All elevation assumptions corresponding to The values ​​are compiled into a table to construct time series statistical dispersion data describing the time stability of the resistance parameters, as shown in Table 2.

[0102] Table 2. Statistics on Riverbed Elevation Assumptions and Roughness Dispersion

[0103]

[0104] See Table 2, which shows the statistical characteristics of the inverted roughness under different riverbed elevation assumptions. The option with the lowest dispersion corresponds to the actual physical state, verifying the parameter locking logic based on time stability.

[0105] S502: Analyze the time series statistical dispersion data and perform a minimum value search, select the riverbed elevation trial sample corresponding to the term with the smallest dispersion, and determine it as the optimal riverbed elevation data;

[0106] Analyzing the dispersion data and executing a minimum search algorithm within the search space, comparing the dispersion indices in Table 2, we find that when the riverbed elevation is assumed to be 12.5 meters, the corresponding roughness sequence dispersion is... The minimum value among all trial samples indicates that, under the assumption that the riverbed elevation is 12.5 meters, the inverted river roughness maintains the best physical stability under seasonal variations of different water levels and flow rates. This is consistent with the objective law that the geological properties of natural river channels do not change drastically in a short period of time. Therefore, the riverbed elevation trial sample of 12.5 meters corresponding to the minimum dispersion term is determined as the optimal riverbed elevation data, that is, the true average riverbed elevation of the river section to be monitored.

[0107] S503: Substitute the optimal riverbed elevation data into the river channel geometric relationship model to lock the model parameters, substitute the water surface width data at the current monitoring time into the locked model for calculation, calculate the cross-sectional area data at the current time, and obtain the area data measurement record.

[0108] Optimal riverbed elevation data Using meters as a fixed parameter, the geometric relationship model of the river channel constructed by S301 is transformed from the original geometric description containing uncertainties into a deterministic functional relationship. For the current monitoring time, the real-time water surface width data measured by S103 is retrieved. meters, if the current water level Given a value (e.g., 25 meters), directly calculate the effective water depth. Meters, if only the width is known, use the calibrated... Find the corresponding water depth value using the relationship curve and apply the definite integral formula. Calculate the cross-sectional area of ​​the water flow, and substitute it into the numerical calculation to obtain the precise cross-sectional area of ​​the water flow at the current moment. The area is measured in square meters, and the value and corresponding timestamp are recorded to obtain the area data calculation record under the condition of no hydrological station, realizing the closed-loop inversion from remote sensing signal to hydrological element.

[0109] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for dynamically calculating the cross-sectional area of ​​river channels without hydrological stations, driven by satellite big data, characterized in that... Includes the following steps: S1: Analyze the multispectral satellite image data of the river section to be monitored and construct the water-land boundary skeleton. Extract the pure water reference spectrum and pure soil reference spectrum along the normal direction of the water-land boundary skeleton, construct a linear hybrid decomposition model, calculate the proportion of sub-pixel level water body, and obtain the water surface width data of the river section. S2: Acquire radar interferometric phase data and river centerline vector data, construct a streamline coordinate system, map the radar interferometric phase data to the streamline coordinate system, set the reverse flow impedance cost to be greater than the forward flow impedance cost, search for the phase path with the minimum cumulative impedance cost, calculate the phase change rate and acquire water surface gradient data. S3: Construct a river channel geometric relationship model, set the search space for riverbed elevation parameters and generate multiple riverbed elevation trial data, combine the water surface width data with the river channel geometric relationship model, calculate the cross-sectional area sequence and hydraulic radius sequence, and obtain the river channel cross-sectional parameter set; S4: Based on the physical balance mechanism between the gravity component of the water flow and the frictional resistance of the riverbed boundary, a fluid resistance inversion model is constructed. The water surface gradient data and the set of geometric parameters of the river cross section are substituted into the fluid resistance inversion logic for reverse deduction. The riverbed resistance coefficient sequence is calculated, and the roughness coefficient data is obtained. S5: Calculate the statistical dispersion of the roughness coefficient data, select the riverbed elevation trial data corresponding to the smallest dispersion term and mark it as the optimal riverbed elevation data, substitute it into the river channel geometric relationship model, calculate the cross-sectional area data, and obtain the area data measurement record.

2. The method for dynamic calculation of river cross-sectional area without hydrological stations driven by satellite big data according to claim 1, characterized in that, The water surface width data includes sub-pixel level river width values ​​and sub-pixel coordinates of the water-land interface; the water surface gradient data includes water surface slope values ​​along the course and flow direction phase gradient components; the river channel cross-section parameter set includes tentative riverbed elevation samples, multiple hypothetical water-passing area matrices, and corresponding hydraulic radius matrices; the roughness coefficient data includes inverted Manning roughness sequence and riverbed comprehensive resistance values; and the area data measurement record includes target time water-passing area values, locked riverbed elevation parameters, and basic information on flow monitoring.

3. The method for dynamic calculation of river cross-sectional area without hydrological stations driven by satellite big data according to claim 1, characterized in that, The specific steps for obtaining the water surface width data are as follows: S101: Acquire multispectral satellite image data of the river section to be monitored, perform band channel spatial convolution processing on the multispectral satellite image data, calculate pixel spatial gradient distribution features and construct image spectral gradient distribution layer, filter local maximum gradient modulus data in the image spectral gradient distribution layer, connect pixel points according to the river flow direction connectivity rules, and construct high gradient boundary skeleton data. S102: Calculate the gradient vector normal direction of the edge pixels in the high gradient boundary skeleton data, search for homogeneous pixels with the minimum gradient modulus in the neighborhood along the normal direction, screen the pure background pixels with the smallest Euclidean distance between the water side and the land side, extract the spectral response features of the pure background pixels, and establish endmember spectral reference data containing pure water spectral features and pure soil spectral features. S103: Call the end-member spectral reference data, analyze the composition of mixed pixel components and construct a linear spectral unmixing model, substitute the multispectral satellite image data into the linear spectral unmixing model to perform sub-pixel decomposition calculation, calculate the water component proportion data in a single pixel, combine the spatial resolution parameters to convert the water component proportion data into entity area and perform horizontal accumulation to obtain the water surface width data of the river cross section.

4. The method for dynamic calculation of river cross-sectional area without hydrological stations driven by satellite big data according to claim 3, characterized in that, The specific steps for obtaining the water surface gradient data are as follows: S201: Obtain radar interferometric phase data and river centerline vector data of the river section to be monitored. Construct a streamline coordinate system with the river centerline as the vertical axis. Calculate the projection coordinates and normal deviation data of the pixels in the radar interferometric phase data relative to the vertical axis. Based on the projection coordinates, resample and map the original phase data to the grid nodes of the streamline coordinate system to establish the streamline system phase mapping matrix. S202: Call the streamline system phase mapping matrix, set the impedance cost of the phase jump in the counter-current direction to be greater than the impedance cost of the phase jump in the downstream direction, search for the pixel connection path with the minimum accumulated impedance cost in the streamline system phase mapping matrix, connect adjacent pixel nodes according to the pixel connection path and smooth the local phase discontinuity region to obtain the flow direction continuous phase sequence. S203: Perform differential calculation along the flow direction for the continuous phase sequence, obtain phase gradient data between adjacent nodes and construct the phase change rate curve along the flow direction, calculate the phase elevation conversion coefficient by combining the radar carrier wavelength parameter and the observation incident angle parameter, and obtain the water surface gradient data of the river section.

5. The method for dynamic calculation of river cross-sectional area without hydrological stations driven by satellite big data according to claim 4, characterized in that, The specific steps for obtaining the river channel cross-section parameter set are as follows: S301: Based on the relationship between the river surface width and the effective water depth, construct a geometric relationship model of the river channel, analyze the range of geographical elevation variation of the river section to be monitored, and set the lower limit and upper limit of the search for the riverbed elevation parameter. Discretize the sampling at fixed intervals between the lower limit and the upper limit of the search to generate a sequence of riverbed elevation trial samples. S302: Call the water surface width data, take each elevation assumption item in the riverbed elevation test sample sequence as the riverbed reference plane, use the river channel geometric relationship model to reverse deduce the effective water depth data for each water surface width, perform geometric integration on the effective water depth data, calculate the water flow area under various elevation assumption conditions, and establish the water flow cross-sectional area matrix. S303: Calculate the wetted perimeter length of each area data in the cross-sectional area matrix corresponding to the water passage, calculate the hydraulic radius data corresponding to each elevation assumption, and associate and combine the multi-assumption cross-sectional area matrix with the hydraulic radius data to obtain the river channel cross-sectional parameter set.

6. The method for dynamically calculating the cross-sectional area of ​​a river channel without hydrological stations, driven by satellite big data, as described in claim 5, is characterized in that... The process of constructing a river channel geometric model based on the relationship between river width and effective water depth is as follows: Based on the hydraulic geometry theory of natural river channel cross-sections, a power function form is selected as the basic mathematical structure for the lateral expansion characteristics of the river channel, and a cross-sectional morphology equation is established with effective water depth as an independent variable and water surface width as a non-independent variable. Geometric integration is performed on the cross-sectional shape equation to derive the formula for calculating the cross-sectional area of ​​the water passage corresponding to the target geometry. Based on the principle of curve arc length integration, a formula for calculating the wetted perimeter is constructed, and an algebraic relationship logic between the cross-sectional area, wetted perimeter, and hydraulic radius is established. The riverbed elevation parameter to be optimized is defined as the vertical starting datum of the cross-sectional geometric profile. The relative positional relationship between the effective water depth and the riverbed elevation parameter is set in the algebraic association logic to construct a full cross-sectional geometric description system containing vertical datum information. For complex riverbed landforms, a piecewise function is used to construct geometric mapping rules for the main channel area and the floodplain area respectively. By splicing the morphological equations of different water level intervals, a parametric channel geometric relationship model that continuously describes the evolution of cross-sectional shape from the dry season to the wet season is generated.

7. The method for dynamically calculating the cross-sectional area of ​​a river channel without hydrological stations, driven by satellite big data, as described in claim 5, is characterized in that... The specific steps for obtaining the roughness coefficient data are as follows: S401: Based on the physical balance mechanism between the gravity component of the water flow and the frictional resistance of the riverbed boundary, the motion law of the open channel fluid under gravity drive is analyzed, the dynamic equations relating the hydraulic radius, cross-sectional area and energy gradient are constructed, the resistance coefficient is set as the dependent variable to be solved and the calculation rules for inferring the resistance characteristics using geometric parameters are established, and a fluid resistance inversion model is constructed. S402: Call the river cross-section parameter set, extract the cross-sectional area sequence and hydraulic radius sequence corresponding to different riverbed elevation assumptions, and perform data association and dimension alignment between the water surface gradient data and the cross-sectional area sequence and hydraulic radius sequence to generate the hydrodynamic calculation input sequence; S403: Substitute the hydrodynamic calculation input sequence into the fluid resistance inversion model for inverse calculation, calculate the theoretical resistance parameters required to maintain the current water flow state for each riverbed elevation trial data, traverse the elevation parameter search space and generate a resistance coefficient distribution sequence corresponding to various riverbed elevation assumptions, and obtain roughness coefficient data.

8. The method for dynamically calculating the cross-sectional area of ​​a river channel without hydrological stations, driven by satellite big data, as described in claim 7, is characterized in that... The process of constructing a fluid resistance inversion model is as follows: Based on the physical balance mechanism between the gravity component of water flow along the channel and the frictional resistance at the riverbed boundary, the motion law of open channel fluid driven by gravity is analyzed, a dynamic equation relating hydraulic radius, cross-sectional area, and energy gradient is constructed, the resistance coefficient is set as the dependent variable to be solved, and a calculation rule for inferring resistance characteristics using geometric parameters is established. The force equilibrium state of the water flow in the monitored river section under the action of gravity is analyzed. The numerical balance relationship between the gravity component along the direction of water flow and the frictional resistance generated by the riverbed boundary is set and used as a physical constraint condition for fluid motion. Based on the physical constraints, a hydraulic control equation describing the uniform flow motion state of an open channel is constructed. A function mapping logic is set that the average flow velocity at the cross section is inversely proportional to the riverbed resistance coefficient and directly proportional to the power function factor of the hydraulic radius and the square root term of the energy gradient. Based on the mass conservation condition of water flow, a continuity equation is constructed in which the instantaneous flow rate of the cross-section is equal to the product of the cross-sectional area and the average flow velocity of the cross-section. The continuity equation is then used to substitute variables in the hydraulic control equation to eliminate the flow velocity variable in the equation. The equations after variable substitution are analyzed in reverse, and the riverbed resistance coefficient is separated into independent terms to be solved. A fluid resistance inversion model is established with the cross-sectional area, the power function factor of the hydraulic radius, the square root term of the energy gradient, and the instantaneous flow rate of the cross-section as calculation factors.

9. The method for dynamically calculating the cross-sectional area of ​​a river channel without hydrological stations, driven by satellite big data, as described in claim 7, is characterized in that... The specific steps for obtaining the area data measurement records are as follows: S501: Perform time-dimensional stability analysis on the roughness coefficient data, calculate the fluctuation amplitude of the resistance coefficient sequence corresponding to various riverbed elevation assumptions in multi-temporal observations, quantitatively evaluate the discrete characteristics of each sequence, and construct time series statistical dispersion data. S502: Analyze the time series statistical dispersion data and perform a minimum value search, select the riverbed elevation trial sample corresponding to the item with the smallest dispersion, and determine it as the optimal riverbed elevation data; S503: Substitute the optimal riverbed elevation data into the river channel geometric relationship model to lock the model parameters, substitute the water surface width data at the current monitoring time into the locked model for calculation, calculate the cross-sectional area data at the current time, and obtain the area data measurement record.

10. The method for dynamic calculation of river cross-sectional area without hydrological stations driven by satellite big data according to claim 9, characterized in that, The process of substituting the optimal riverbed elevation data into the river channel geometric relationship model to lock the model parameters, and then substituting the water surface width data at the current monitoring time into the locked model for calculation to determine the cross-sectional area data at the current time is as follows: The optimal riverbed elevation data is set as the vertical starting point of the channel geometric relationship model. It is substituted into the channel geometric relationship model to determine the zero point of the effective water depth variable, and the numerical locking of the undetermined geometric parameters in the model is performed to generate a unique and definite cross-sectional shape equation for the river section to be monitored. The water surface width data at the current monitoring time is called up, and based on the function characteristic of the water surface width changing monotonically with the effective water depth in the cross-sectional shape equation, the real-time effective water depth value corresponding to the current water surface width value is calculated. An integral operation rule with effective water depth as the independent variable is set, the lower limit of the integration interval is set to zero water depth, and the upper limit is set to the real-time effective water depth value. A definite integral calculation is performed on the cross-sectional shape equation to obtain the geometric area of ​​the cross-section under the current water level conditions, and the cross-sectional area data is output.

Images