A method for analyzing nonlinear runoff characteristics of small watersheds in valley terrain

Abstract

The present invention discloses a method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain, comprising: S1: obtaining valley terrain-hydrological data and spatially dividing hydrological grid units; S2: identifying hydrological behavior characteristics of hydrological grid units to form independent hydrological response units; S3: analyzing rainfall infiltration in the hydrological response unit to form a first influence degree, surface runoff to form a second influence degree, and soil moisture dynamic changes to form a third influence degree, and constructing a valley dynamic fluid model; S4: combining the valley dynamic fluid model with a groundwater flow model and a surface water flow model to form a hydrological-hydrodynamic simulation framework; S5: using the hydrological-hydrodynamic simulation framework to simulate various rainfall scenarios, combining valley micro-topography factors for analysis, calculating the flow change rate of different rainfall scenarios, and identifying nonlinear characteristics. The present invention realizes high-precision simulation and prediction of small watersheds in valley terrain, and improves the accuracy and timeliness of runoff analysis.

Description

Technical Field

[0001] The invention relates to the technical field of valley hydrological simulation, and in particular to a method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain. Background Art

[0002] The characteristics of flood runoff in very small watersheds are characterized by high flood peaks, large volumes, steep rises and falls, etc. Accurately understanding and predicting the runoff characteristics in valley terrain watersheds is crucial for flood prevention and disaster reduction, rational allocation of water resources, and ecological environmental protection. However, due to the complex terrain and small catchment area of ​​valley terrain small watersheds, rainfall events often lead to rapid and drastic surface runoff reactions. Traditional linear hydrological models are difficult to accurately simulate the hydrological processes in these areas, especially in terms of nonlinear runoff characteristics. Therefore, it is urgent to develop a new analysis method that can effectively deal with the nonlinear runoff characteristics of valley terrain small watersheds.

[0003] Traditional technical solutions mainly rely on empirical formulas and statistical analysis to predict runoff characteristics, such as the use of unit line theory, hydrological response function and other methods. Although these methods can provide useful prediction results to a certain extent, their prediction accuracy is often unsatisfactory when faced with nonlinear hydrological processes under complex terrain conditions. In addition, traditional methods usually ignore the continuity in the time dimension and the heterogeneity in the spatial dimension, which leads to a further decline in the model's prediction ability. Existing technologies attempt to improve this situation by introducing more complex mathematical models, such as using finite element method or finite difference method for numerical simulation, but these methods have high computational costs and strict requirements on initial conditions and boundary conditions, and have great limitations in practical applications.

[0004] This technical solution proposes a method based on adaptive grid division, improved clustering algorithm to identify hydrological response units, distributed hydrological model establishment and groundwater flow simulation to more accurately analyze the nonlinear runoff characteristics of valley terrain small watersheds, and provide a scientific basis for flood prediction, risk prevention and control, and reservoir water level regulation in valley-type small watersheds. Summary of the invention

[0005] The present invention provides a method for analyzing nonlinear runoff generation and confluence characteristics of a small watershed in a valley terrain, comprising:

[0006] S1: Obtain valley topography-hydrological data through digital elevation model and valley terrain coverage data, use adaptive grid technology to spatially divide hydrological grid units, and obtain terrain characteristics, soil properties and vegetation coverage within the hydrological grid units;

[0007] S2: The hydrological grid cells are processed by clustering algorithms to identify areas with hydrological behavior characteristics, which are defined as independent hydrological response units;

[0008] S3: Analyze the rainfall infiltration in the hydrological response unit to form the first impact, analyze the surface runoff to form the second impact, analyze the dynamic change of soil moisture to form the third impact, and construct a valley dynamic fluid model by integrating the hydrological processes of the first impact, the second impact and the third impact;

[0009] S4: Combine the valley dynamic fluid model with the groundwater flow model and the surface water flow model to form a hydrological-hydrodynamic simulation framework;

[0010] S5: Use the hydrological-hydrodynamic simulation framework to simulate various rainfall scenarios, identify key factors of surface runoff, analyze valley microtopography factors, determine peak flow and frequency, calculate the flow change rate of different rainfall scenarios, and identify nonlinear characteristics.

[0011] Preferably, the spatial division of hydrological grid units by using the adaptive grid technology in S1 comprises the following steps:

[0012] S1.1. The soil types and vegetation coverage obtained from digital elevation models and satellite remote sensing images form a multivariate geographic information data set. The multivariate geographic information data is unified into a coordinate system using spatial data fusion technology, and noise is removed and smoothed.

[0013] S1.2. Use convolutional neural networks to identify and evaluate the terrain complexity and hydrological response differences of each point in the study area, and intelligently adjust the grid size and boundaries according to factors such as terrain slope, soil permeability, and vegetation coverage, so that a dense grid layout is formed in areas with rapid changes in surface runoff or active groundwater recharge, and a sparse grid is formed in uniform areas to achieve spatial discretization;

[0014] S1.3. After completing the preliminary grid division, the influence of each grid unit on the overall hydrological process is evaluated by combining the key indicators of water flow path and waterlogging area distribution in the preliminary simulation results. For specific areas with significant nonlinear response characteristics or sensitivity to flood events, the grid scale is further reduced and local details are increased.

[0015] Preferably, in said S1.1, the multivariate geographic information data is unified into a coordinate system by using spatial data fusion technology, and the coordinate transformation is performed by using a spatial transformation model of the least squares method through the acquired multivariate geographic information data set to align the data in the preset geographic coordinates;

[0016] Defining a multivariate geographic information dataset and the target coordinate system The conversion relationship between them is: ,in, is the matrix of rotation and translation parameters, is the translation vector, is the data coordinate point; by selecting multiple data coordinate points and using the least squares method to solve the best and Value, coordinate conversion; the multivariate geographic information dataset The fusion between the data layers is carried out through spatial data fusion technology, and the formula is: ,in is the value of the fused data coordinate point, For the Data layers at data coordinate points The value at is the total number of data layers, is the data layer weight coefficient; the spatial unification of multi-source geographic information data is achieved through coordinate transformation and fusion.

[0017] Preferably, the S1.2 uses a convolutional neural network to identify and evaluate the terrain complexity and potential hydrological response differences of each point in the study area, and intelligently adjusts the grid size and boundaries, including the following:

[0018] S1.2.1, using digital elevation model, soil permeability and vegetation coverage as input layers;

[0019] S1.2.2. The spatial features are extracted through the convolutional layers of the convolutional neural network model. Each convolutional layer is represented as in It is The input feature map is at position The value at is the convolution kernel weight, is the bias term, Represents the convolution operation;

[0020] S1.2.3. Reduce feature dimensions through the pooling layer of the convolutional neural network model;

[0021] S1.2.4. Defining the terrain complexity index output by the convolutional neural network model , the formula is: ,in Is the last convolutional or fully connected layer at position The feature map output, is the importance weight of the feature map, is a nonlinear activation function, is a feature atlas;

[0022] S1.2.5. Based on the terrain complexity index The grid size and boundary are adjusted dynamically so that the complex area has dense grid division, while the flat area uses sparse grid to form the discretization of valley basin space.

[0023] Preferably, the clustering algorithm extracts a series of hydrological characteristic parameters from each hydrological grid unit, including but not limited to terrain slope, soil permeability, and vegetation coverage, and calculates the distance metric between different grids through the clustering algorithm to determine the similarity of the grid units. The formula is: ,in and Respectively represent and The grid cell in The value of the hydrological characteristic, It is The weight coefficient of each feature reflects its importance in distinguishing hydrological behaviors. is the distance metric, is the number of grid cells; The hydrological grid cells are divided into a plurality of clusters, and the grid cells in each cluster exhibit highly similar hydrological behavior characteristics, and the clusters of highly similar hydrological behavior characteristics are defined as independent hydrological response units.

[0024] Preferably, the construction of the valley dynamic fluid model forms a first influence degree by simulating the rainfall infiltration process of the hydrological response unit, the influence of soil properties, vegetation coverage and terrain slope on the infiltration rate; the first influence degree reflects the change in the amount of groundwater converted from rainfall under different conditions; the surface runoff path is analyzed through the spatial data of the hydrological response unit, and the convergence and dispersion of water on the surface are evaluated in combination with the terrain and land use conditions to form a second influence degree, which describes the surface water flow pattern and its contribution to local flood events; the dynamic changes of soil moisture in the hydrological response unit are analyzed, and the storage, transmission and evaporation process of moisture in the soil are tracked to form a third influence degree, which captures the changes in soil moisture status over time and its impact on subsequent hydrological processes; the first to third influence degrees are integrated to form a valley dynamic fluid model.

[0025] Preferably, the hydrological-hydrodynamic simulation framework in S4 is formed by combining a valley dynamic fluid model with a groundwater flow model and a surface water flow model; the groundwater flow model is formed by a first influence degree; and the surface water flow model is formed by a second influence degree.

[0026] The groundwater flow model uses the three-dimensional groundwater flow equation ,in is the permeability tensor, is the height of groundwater level change, is the specific storage coefficient, is the source-sink term;

[0027] The surface water flow model uses the Saint-Venant equations to describe the surface runoff process, and the formula is: ,here is the cross-sectional area of ​​water flow, It's traffic. is the distance along the river channel, is the net rain intensity.

[0028] Preferably, the hydrological-hydrodynamic simulation framework is formed by combining a valley dynamic fluid model with a groundwater flow model and a surface water flow model, and the formula is: ,in is the surface elevation, is the surface runoff flow, is the exchange rate, is the height of groundwater level change, and are the weight coefficients of the exchange rate between groundwater and surface water, respectively, which are used to quantify the interaction strength between the two.

[0029] Preferably, S5 comprises the following steps:

[0030] S5.1. Use the hydrological-hydrodynamic simulation framework to simulate rainfall scenarios of varying intensity and duration.

[0031] S5.2. Identify the surface runoff response parameters under different rainfall scenarios and extract the key factors affecting the surface runoff behavior;

[0032] S5.3. Combine valley micro-topography characteristic data to analyze the impact of valley topography on water flow paths and convergence points, and determine peak flood flow and frequency;

[0033] S5.4. Calculate the rate of change of flow per unit time through time series data under different rainfall scenarios, and identify mutation points and periodic fluctuations in flow changes.

[0034] Preferably, the peak flow and frequency in S5.3 are obtained by comparing the simulation results with the changes in runoff flow under different rainfall events, identifying the conditions leading to the formation of peaks, and calculating the extreme peak flow and the probability of peak occurrence using extreme value theory.

[0035] Compared with the prior art, the technical solution of this application has the following technical effects:

[0036] The present invention combines the digital elevation model with valley terrain coverage data and uses adaptive grid technology to spatially divide hydrological grid units. This solution can obtain accurate terrain characteristics, soil properties and vegetation coverage information, solving the problem of difficulty in accurately reflecting complex terrain and heterogeneous surface conditions in traditional watershed analysis, and improving the accuracy of understanding the nonlinear runoff characteristics of very small watersheds. The constructed hydrological grid units provide a reliable foundation for subsequent hydrological simulations, making the simulation results closer to the actual situation.

[0037] The present invention uses a clustering algorithm to identify areas with specific hydrological behavior characteristics and defines them as independent hydrological response units, solving the technical problem of how to effectively distinguish different hydrological response sections in large-scale data analysis. By combining static factors such as terrain slope and soil permeability and dynamic changes in vegetation coverage, the hydrological process in the basin can be finely portrayed. This definition of hydrological response units based on feature clustering helps to deepen the understanding of the water cycle mechanism within the basin, and at the same time provides strong support for establishing a more accurate hydrological-hydrodynamic simulation framework.

[0038] The present invention constructs a valley dynamic fluid model by integrating the first influence degree formed by rainfall infiltration, the second influence degree formed by surface runoff and the third influence degree formed by dynamic changes in soil moisture. This solves the problem that previous models cannot fully consider the interaction of multiple hydrological processes. It can comprehensively evaluate the changes in groundwater volume under rainfall conditions, surface water flow patterns and their contribution to local flood events, and changes in soil moisture status over time. It can predict the location of water flow paths and convergence points under different rainfall scenarios, and determine peak flow and frequency.

[0039] The present invention solves the challenge of cross-media water flow exchange simulation by combining the valley dynamic fluid model with the groundwater flow model and the surface water flow model to form a hydrological-hydrodynamic simulation framework, allowing the interaction between groundwater and surface water to be simulated on a unified platform, using three-dimensional groundwater flow equations and Saint-Venant equations to describe their respective processes, and improving the accuracy of the model by quantifying the intensity of the interaction between the two, improving the understanding of complex hydrological systems, and enhancing the ability to simulate flow changes under different rainfall scenarios, including the identification of mutation points and periodic fluctuations. Combined with micro-topography factors, the present invention forms a timely and accurate flood assessment of valley terrain, and realizes an efficient and accurate analysis method of runoff generation and confluence characteristics of ultra-small watersheds in valley areas.

[0040] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application so that it can be implemented in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following is a detailed description of the preferred embodiments of the present application in conjunction with the accompanying drawings as follows.

[0041] Based on the detailed description of the specific embodiments of the present application in combination with the accompanying drawings below, those skilled in the art will become more aware of the above and other objects, advantages and features of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings without creative work. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale.

[0043] Figure 1 This is a flow chart of a nonlinear runoff generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0044] Figure 2 A grid division flow chart of a nonlinear runoff generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0045] Figure 3 A grid adjustment flow chart of a nonlinear runoff generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0046] Figure 4 It is a simulation flow chart of a nonlinear runoff generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0047] Figure 5 A grid diagram of a nonlinear runoff generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0048] Figure 6 A valley dynamic fluid model structure diagram of a nonlinear flow generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0049] Figure 7 A 24-hour runoff intensity diagram of a nonlinear runoff generation and confluence analysis method for a small watershed in a valley terrain according to the present invention;

[0050] Figure 8 The present invention discloses a peak flow diagram of a nonlinear flow generation and confluence analysis method for a small watershed in a valley terrain. DETAILED DESCRIPTION

[0051] To make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiment of the present application. Obviously, the described embodiment is a part of the embodiment of the present application, rather than all of the embodiments. In the following description, specific details such as specific configuration and components are provided only to help fully understand the embodiments of the present application. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. In addition, for clarity and brevity, the description of known functions and structures is omitted in the embodiment.

[0052] It should be understood that the references to "one embodiment" or "this embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present application. Therefore, the references to "one embodiment" or "this embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0053] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity, and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0054] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" in this article describes another type of association object relationship, indicating that there can be two relationships. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the previous and next associated objects are in an "or" relationship.

[0055] The term "at least one" in this article is merely a description of the association relationship of associated objects, indicating that there may be three relationships. For example, at least one of A and B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0056] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusions.

[0057] Example 1

[0058] This embodiment mainly describes a method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain. Figure 1 As shown, including;

[0059] S1: Obtain valley topography-hydrological data through digital elevation model and valley terrain coverage data, use adaptive grid technology to spatially divide hydrological grid units, and obtain terrain characteristics, soil properties and vegetation coverage within the hydrological grid units;

[0060] S2: The hydrological grid cells are processed by clustering algorithms to identify areas with hydrological behavior characteristics, which are defined as independent hydrological response units;

[0061] S3: Analyze the rainfall infiltration in the hydrological response unit to form the first impact, analyze the surface runoff to form the second impact, analyze the dynamic change of soil moisture to form the third impact, and construct a valley dynamic fluid model by integrating the hydrological processes of the first impact, the second impact and the third impact;

[0062] S4: Combine the valley dynamic fluid model with the groundwater flow model and the surface water flow model to form a hydrological-hydrodynamic simulation framework;

[0063] S5: Use the hydrological-hydrodynamic simulation framework to simulate various rainfall scenarios, identify key factors of surface runoff, analyze valley microtopography factors, determine peak flow and frequency, calculate the flow change rate of different rainfall scenarios, and identify nonlinear characteristics.

[0064] Further, if Figure 2 As shown, the spatial division of hydrological grid units using adaptive grid technology in S1 includes the following steps:

[0065] S1.1. The soil types and vegetation coverage obtained from digital elevation models and satellite remote sensing images form a multivariate geographic information data set. The multivariate geographic information data is unified into a coordinate system using spatial data fusion technology, and noise is removed and smoothed.

[0066] S1.2. Use convolutional neural networks to identify and evaluate the terrain complexity and hydrological response differences of each point in the study area, and intelligently adjust the grid size and boundaries according to factors such as terrain slope, soil permeability, and vegetation coverage, so that a dense grid layout is formed in areas with rapid changes in surface runoff or active groundwater recharge, and a sparse grid is formed in uniform areas to achieve spatial discretization;

[0067] S1.3. After completing the preliminary grid division, the influence of each grid unit on the overall hydrological process is evaluated by combining the key indicators of water flow path and waterlogging area distribution in the preliminary simulation results. For specific areas with significant nonlinear response characteristics or sensitivity to flood events, the grid scale is further reduced and local details are increased.

[0068] Furthermore, in S1.1, the multivariate geographic information data is unified into a coordinate system through spatial data fusion technology, and the spatial transformation model of the least squares method is used to transform the coordinates through the acquired multivariate geographic information data set, so as to align the data in the preset geographic coordinates;

[0069] Defining a multivariate geographic information dataset and the target coordinate system The conversion relationship between them is: ,in, is the matrix of rotation and translation parameters, is the translation vector, is the data coordinate point; by selecting multiple data coordinate points and using the least squares method to solve the best and Value, coordinate conversion; multivariate geographic information dataset The fusion between the data layers is carried out through spatial data fusion technology, and the formula is: ,in is the value of the fused data coordinate point, For the Data layers at data coordinate points The value at is the total number of data layers, is the data layer weight coefficient; spatial unification of multi-source geographic information data is achieved through coordinate transformation and fusion.

[0070] Further, if Figure 3 As shown in the figure, the terrain complexity and potential hydrological response differences of each point in the study area are identified and evaluated through a convolutional neural network, and the grid size and boundaries are intelligently adjusted, including the following steps:

[0071] S1.2.1, using digital elevation model, soil permeability and vegetation coverage as input layers;

[0072] S1.2.2. The spatial features are extracted through the convolutional layers of the convolutional neural network model. Each convolutional layer is represented as in It is The input feature map is at position The value at is the convolution kernel weight, is the bias term, Represents the convolution operation;

[0073] S1.2.3. Reduce feature dimensions through the pooling layer of the convolutional neural network model;

[0074] S1.2.4. Defining the terrain complexity index output by the convolutional neural network model , the formula is: ,in Is the last convolutional or fully connected layer at position The feature map output, is the importance weight of the feature map, is a nonlinear activation function, is a feature atlas;

[0075] S1.2.5. Based on the terrain complexity index Dynamically adjust the grid size and boundaries so that complex areas have dense grid divisions, while flat areas use sparse grids to form a discretization of the valley basin space;

[0076] According to the complexity index The output layer outputs high complexity, medium complexity and low complexity, and then divides the terrain grid into high and low complexity forms.

[0077] Furthermore, the clustering algorithm extracts a series of hydrological characteristic parameters from each hydrological grid unit, including but not limited to terrain slope, soil permeability, and vegetation coverage. The clustering algorithm is used to calculate the distance metric between different grids and judge the similarity of grid units. The formula is: ,in and Respectively represent and The grid cell in The value of the hydrological characteristic, It is The weight coefficient of each feature reflects its importance in distinguishing hydrological behaviors. is the distance metric, is the number of grid cells; according to the distance metric The hydrological grid cells are divided into multiple clusters. The grid cells in each cluster show highly similar hydrological behavior characteristics. Clusters with highly similar hydrological behavior characteristics are defined as independent hydrological response units.

[0078] Furthermore, the construction of the valley dynamic fluid model forms the first influence dimension by simulating the rainfall infiltration process of the hydrological response unit, the influence of soil properties, vegetation coverage and terrain slope on the infiltration rate; the first influence dimension reflects the change in the amount of groundwater converted from rainfall under different conditions; the surface runoff path is analyzed through the spatial data of the hydrological response unit, and the convergence and dispersion of water on the surface are evaluated in combination with the terrain and land use conditions to form the second influence dimension, which describes the surface water flow pattern and its contribution to local flood events; the dynamic changes of soil moisture in the hydrological response unit are analyzed, and the storage, transmission and evaporation of moisture in the soil are tracked to form the third influence dimension, which captures the changes in soil moisture status over time and its impact on subsequent hydrological processes; the first to third influence dimensions are integrated to form the valley dynamic fluid model.

[0079] Furthermore, the hydrological-hydrodynamic simulation framework in S4 is formed by combining the valley dynamic fluid model with the groundwater flow model and the surface water flow model; the groundwater flow model is formed by the first influence degree; and the surface water flow model is formed by the second influence degree.

[0080] The groundwater flow model uses the three-dimensional groundwater flow equation ,in is the permeability tensor, is the groundwater level, is the specific storage coefficient, is the source-sink term;

[0081] The surface water flow model uses the Saint-Venant equations to describe the surface runoff process, and the formula is: ,here is the cross-sectional area of ​​water flow, It's traffic. is the distance along the river channel, is the net rain intensity.

[0082] Furthermore, the hydrological-hydrodynamic simulation framework is formed by combining the valley dynamic fluid model with the groundwater flow model and the surface water flow model, and the formula is: ,in is the surface elevation, is the surface runoff flow, is the exchange rate, and are the weight coefficients of the exchange rate between groundwater and surface water, respectively, which are used to quantify the interaction strength between the two.

[0083] Further, if Figure 4 As shown, S5 includes the following steps:

[0084] S5.1. Use the hydrological-hydrodynamic simulation framework to simulate rainfall scenarios of varying intensity and duration.

[0085] S5.2. Identify the surface runoff response parameters under different rainfall scenarios and extract the key factors affecting the surface runoff behavior;

[0086] S5.3. Combine valley micro-topography characteristic data to analyze the impact of valley topography on water flow paths and convergence points, and determine peak flood flow and frequency;

[0087] S5.4. Calculate the rate of change of flow per unit time through time series data under different rainfall scenarios, and identify mutation points and periodic fluctuations in flow changes.

[0088] Furthermore, the peak flow and frequency in S5.3 are compared with the simulation results to identify the changes in runoff flow under different rainfall events, identify the conditions that lead to the formation of peaks, and use extreme value theory to calculate the extreme peak flow and the probability of peak occurrence.

[0089] This embodiment combines the digital elevation model with the valley terrain coverage data, and uses the adaptive grid technology to spatially divide the hydrological grid units to obtain accurate terrain characteristics, soil properties and vegetation coverage information. The clustering algorithm is used to identify areas with specific hydrological behavior characteristics and define them as independent hydrological response units. The static factors of terrain slope and soil permeability and the dynamic vegetation coverage changes are combined to establish a more accurate hydrological-hydrodynamic simulation framework. The valley dynamic fluid model is constructed by integrating the first influence degree formed by rainfall infiltration, the second influence degree formed by surface runoff, and the third influence degree formed by the dynamic changes in soil moisture. The groundwater changes under rainfall conditions, the surface water flow pattern and its contribution to local flood events, and the changes in soil moisture status over time are comprehensively evaluated. The water flow path and the location of the convergence point are predicted under different rainfall scenarios, the peak flow and frequency are determined, and the nonlinear runoff generation and convergence characteristics of the valley terrain ultra-small watershed are analyzed.

[0090] Example 2

[0091] This embodiment describes in detail a specific characteristic analysis method for nonlinear runoff generation and convergence in a small watershed with a valley terrain, specifically including:

[0092] Select a valley terrain, obtain valley terrain-hydrological data through digital elevation model and valley terrain coverage data, and use adaptive grid technology to spatially divide hydrological grid units; obtain terrain characteristics, soil properties and vegetation coverage within the hydrological grid units;

[0093] The grid size and boundary are dynamically adjusted according to the valley terrain complexity index, so that the complex area has dense grid division, while the flat area uses sparse grid, forming the discretization of the valley basin space, such as Figure 5 As shown, the valley is divided into multi-region modules according to the boundaries, and the confluence flow path is formed in the multi-region modules according to the valley terrain.

[0094] like Figure 6 As shown in the figure, by simulating the rainfall infiltration process of the hydrological response unit, the influence of soil properties, vegetation coverage and terrain slope on the infiltration rate is formed to form the first influence degree; the first influence degree reflects the change of rainfall converted into groundwater under different conditions;

[0095] Analyze the surface runoff path through the spatial data of the hydrological response unit, and evaluate the convergence and dispersion of water on the surface in combination with the topography and land use to form the second impact factor, which describes the surface water flow pattern and its contribution to local flood events;

[0096] Analyze the dynamic changes of soil moisture in the hydrological response unit, track the storage, transmission and evaporation of moisture in the soil, and form a third impact factor that captures the changes in soil moisture status over time and its impact on subsequent hydrological processes;

[0097] The first to third influence degrees are integrated to form a valley dynamic fluid model;

[0098] The hydrological-hydrodynamic simulation framework is formed by combining the valley dynamic fluid model with the groundwater flow model and the surface water flow model, and the hydrological-hydrodynamic simulation framework is used to simulate rainfall scenarios covering different intensities and durations;

[0099] according to Figure 5 The valley terrain grid map shown in the figure selects point A to simulate three rainfall scenarios. Point A is selected as the intersection of the runoff paths to better evaluate the runoff characteristics under different conditions. The three rainfall scenarios are short-term heavy rainfall, long-term weak rainfall, and continuous multiple rainfall.

[0100] The simulation is based on a hydrological-hydrodynamic simulation framework that combines a valley dynamic fluid model with a groundwater flow model and a surface water flow model; concentrated grid division is performed for complex areas (such as riverbeds and steep slopes), and sparse grid division is performed for simple areas (such as flat highlands);

[0101] Point A was selected to generate valley flow by simulating the first and third impacts through the hydrological-hydrodynamic simulation framework for short-term heavy rainfall, long-term weak rainfall, and continuous multiple rainfall;

[0102] Point A was selected to simulate the second impact degree and generate valley runoff velocity through the hydrological-hydrodynamic simulation framework for short-term heavy rainfall, long-term weak rainfall and continuous multiple rainfall;

[0103] The second influence degree is used to generate complex regional grids and simple regional grids combined with valley micro-topography feature data to obtain the impact of valley topography on water flow paths and convergence points;

[0104] like Figure 7 As shown in Figure 1, by combining the valley flow rate, valley runoff velocity and micro-topography characteristics of point A, and observing for one day, a 24-hour runoff intensity curve of short-term heavy rainfall, long-term weak rainfall and continuous multiple rainfall based on the hydrological-hydrodynamic simulation framework is generated. Figure 7 It can be found that for the selected location of point A, during the short-term heavy rainfall, the infiltration rate of the first impact degree is high, the soil moisture state in the third impact degree is low, and the surface water flow of the second impact degree is accelerated. Therefore, in the observation in the early part of the day, the first flood peak is quickly ushered in during the short-term heavy rainfall, the flood peak flow is high, and the time is short. In the observation in the middle of the day, due to the low infiltration rate and the high soil moisture state in the third impact degree, the surface water flow of the second impact degree is accelerated, and the second short-term heavy rainfall ushered in the second flood peak, and the flood peak flow continued to increase and the time increased;

[0105] For the long-term weak rainfall process, the infiltration rate of the first impact degree is higher, the soil moisture state in the third impact degree is lower, and the surface water flow of the second impact degree is slower. The peak flow rate decreases during the long-term weak rainfall process. In the early part of the day, the peak flow rate decreases during the long-term weak rainfall process, and the time is shorter. In the middle part of the day, the flow rate continues to decrease during the long-term weak rainfall process, and the time is longer. In the later part of the day, due to the long-term weak rainfall, the infiltration rate of the first impact degree is lower, the soil moisture state in the third impact degree is higher, and the surface water flow of the second impact degree is faster, resulting in a lower flow rate and a longer time.

[0106] For consecutive rainfalls, the infiltration rate of the first impact degree is higher, the soil moisture state in the third impact degree is lower, and the surface water flow of the second impact degree is slower. In the first half of the day, the peak flow increases during consecutive rainfalls, and the time is shorter. In the middle of the day, due to consecutive rainfalls, the infiltration rate of the first impact degree is higher, the soil moisture state in the third impact degree is higher, and the surface water flow of the second impact degree is faster. During consecutive rainfalls, the flow continues to increase and the time is longer. In the second half of the day, due to consecutive rainfalls, the infiltration rate of the first impact degree is higher, the soil moisture state in the third impact degree is higher, and the surface water flow of the second impact degree is faster, resulting in multiple peaks in the flow, and the time is shorter.

[0107] Through the analysis of the above-mentioned 24-hour runoff intensity diagram, the present invention utilizes the hydrological-hydrodynamic simulation framework to simulate the peak flow of short-term heavy rainfall, long-term weak rainfall and continuous multiple rainfall in a specific valley basin, and can accurately analyze the runoff generation and confluence analysis of the specific valley basin.

[0108] like Figure 8 As shown in the figure, a peak flow diagram is formed based on the flow monitoring at point A, which shows the flow changes under three different rainfall patterns. The horizontal axis in the figure represents time (hours), and the vertical axis represents the peak flow (cubic meters / second).

[0109] In the case of short-term heavy rainfall, the peak flow rate rises rapidly, reaching a maximum value of nearly 40 cubic meters per second in about 8 hours, and then the flow rate drops rapidly, returning to a lower level after about 12 hours.

[0110] In the case of long-term weak rainfall, the peak flow rate rises and falls slowly, reaching a maximum value of about 15 cubic meters per second in about 10 hours, and then the flow rate gradually decreases, returning to a lower level after about 15 hours.

[0111] In the case of multiple consecutive rainfalls, the changes in peak flood flow are more complicated. The flow shows multiple fluctuations under the influence of multiple rainfalls. The maximum value occurs at around 9 hours, close to 25 cubic meters per second, and then the flow gradually decreases and returns to a lower level after about 15 hours.

[0112] In general, the peak flow caused by short-term heavy rainfall changes most dramatically, while the flow caused by long-term weak rainfall changes more slowly. Continuous rainfall leads to multiple fluctuations in flow.

[0113] This embodiment integrates the first influence degree, the second influence degree and the third influence degree to form a valley dynamic fluid model, and uses the valley dynamic fluid model in combination with a groundwater flow model and a surface water flow model to form a hydrological-hydrodynamic simulation framework to simulate the flow changes under different rainfall scenarios, including the identification of mutation points and periodic fluctuations. Combined with micro-topography factors, the timeliness and accuracy of flood assessment of valley terrain are formed, and an efficient and accurate analysis method for runoff generation and confluence characteristics of ultra-small watersheds in valley areas is implemented.

[0114] The above are only preferred embodiments of the present invention, which do not limit the scope of protection of the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any changes, modifications, replacements, integrations and parameter changes to these embodiments within the spirit and principles of the present invention through conventional substitutions or without departing from the principles and spirit of the present invention fall within the scope of protection of the present invention.

Claims

1. A method for analyzing nonlinear runoff characteristics of small watersheds in valley terrain, characterized in that: include; S1: Obtain valley terrain-hydrological data through digital elevation model and valley terrain coverage data, use adaptive grid technology to spatially divide hydrological grid units, obtain terrain features within hydrological grid units, extract land use and vegetation coverage, and analyze soil texture; S2: The hydrological grid cells are processed by clustering algorithms to identify areas with hydrological behavior characteristics, which are defined as independent hydrological response units; S3: Analyze the rainfall infiltration in the hydrological response unit to form the first impact, analyze the surface runoff to form the second impact, analyze the dynamic change of soil moisture to form the third impact, and construct a valley dynamic fluid model by integrating the hydrological processes of the first impact, the second impact and the third impact; S4: Combine the valley dynamic fluid model with the groundwater flow model and the surface water flow model to form a hydrological-hydrodynamic simulation framework; S5: Use the hydrological-hydrodynamic simulation framework to simulate various rainfall scenarios, identify key factors of surface runoff, analyze valley microtopography factors, determine peak flow and frequency, calculate the flow change rate of different rainfall scenarios, and identify nonlinear characteristics.

2. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 1 is characterized in that: The spatial division of hydrological grid units using the adaptive grid technology in S1 includes the following steps: S1.

1. The soil types and vegetation coverage obtained from digital elevation models and satellite remote sensing images form a multivariate geographic information data set. The multivariate geographic information data is unified into a coordinate system using spatial data fusion technology, and noise is removed and smoothed. S1.

2. Use convolutional neural networks to identify and evaluate the terrain complexity and hydrological response differences of each point in the study area, and intelligently adjust the grid size and boundaries according to factors such as terrain slope, soil permeability, and vegetation coverage, so that a dense grid layout is formed in areas with rapid changes in surface runoff or active groundwater recharge, and a sparse grid is formed in uniform areas to achieve spatial discretization; S1.

3. After completing the preliminary grid division, the key indicators of the valley terrain small watershed in the preliminary simulation results, such as the runoff velocity, flood flow, peak flood time, and catchment area distribution, are combined to evaluate the impact of each grid unit on the overall hydrological process. For specific areas with significant nonlinear response characteristics or sensitivity to flood events, the grid scale is further reduced and local details are increased.

3. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 2 is characterized in that: In S1.1, the multivariate geographic information data is unified into a coordinate system by using spatial data fusion technology, and the coordinate transformation is performed by using the spatial transformation model of the least squares method through the acquired multivariate geographic information data set to align the data in the preset geographic coordinates; Defining a multivariate geographic information dataset and the target coordinate system The conversion relationship between them is: ,in, is the matrix of rotation and translation parameters, is the translation vector, is the data coordinate point; by selecting multiple data coordinate points and using the least squares method to solve the best and Value, coordinate conversion; the multivariate geographic information dataset The fusion between the data layers is carried out through spatial data fusion technology, and the formula is: ,in is the value of the fused data coordinate point, For the Data layers at data coordinate points The value at is the total number of data layers, is the data layer weight coefficient; the spatial unification of multi-source geographic information data is achieved through coordinate transformation and fusion.

4. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 2 is characterized in that: In S1.2, a convolutional neural network is used to identify and evaluate the terrain complexity and potential hydrological response differences of each point in the study area, and the grid size and boundaries are intelligently adjusted, including the following: S1.2.

1. Use digital elevation model, soil permeability and vegetation coverage as input layers; S1.2.

2. The spatial features are extracted through the convolutional layers of the convolutional neural network model. Each convolutional layer is represented as ,in It is The input feature map is at position The value at is the convolution kernel weight, is the bias term, Represents the convolution operation; S1.2.

3. Reduce feature dimensions through the pooling layer of the convolutional neural network model; S1.2.

4. Defining the terrain complexity index output by the convolutional neural network model , the formula is: ,in Is the last convolutional or fully connected layer at position The feature map output, is the importance weight of the feature map, is a nonlinear activation function, is the feature atlas; S1.2.

5. Based on the terrain complexity index The grid size and boundary are adjusted dynamically so that the complex area has dense grid division, while the flat area uses sparse grid to form the discretization of valley basin space.

5. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 1 is characterized in that: The clustering algorithm extracts a series of hydrological characteristic parameters from each hydrological grid unit, including but not limited to terrain slope, soil permeability, and vegetation coverage. The clustering algorithm is used to calculate the distance metric between different grids and judge the similarity of grid units. The formula is: ,in and Respectively represent and The grid cell in The value of the hydrological characteristic, It is The weight coefficient of each feature reflects its importance in distinguishing hydrological behaviors. is the distance metric, is the number of grid cells; the distance metric The hydrological grid cells are divided into a plurality of clusters, and the grid cells in each cluster exhibit highly similar hydrological behavior characteristics, and the clusters of highly similar hydrological behavior characteristics are defined as independent hydrological response units.

6. A method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 1 or 5, characterized in that: The construction of the valley dynamic fluid model forms a first influence degree by simulating the rainfall infiltration process of the hydrological response unit, the influence of soil properties, vegetation coverage and terrain slope on the infiltration rate; the first influence degree reflects the change in the amount of groundwater converted from rainfall under different conditions; the surface runoff path is analyzed through the spatial data of the hydrological response unit, and the convergence and dispersion of water on the surface are evaluated in combination with the terrain and land use conditions to form a second influence degree, which describes the surface water flow pattern and its contribution to local flood events; the dynamic changes of soil moisture in the hydrological response unit are analyzed, and the storage, transmission and evaporation process of moisture in the soil are tracked to form a third influence degree, which captures the changes in soil moisture status over time and its impact on subsequent hydrological processes; the first to third influence degrees are integrated to form a valley dynamic fluid model.

7. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 1 is characterized in that: The hydrological-hydrodynamic simulation framework in S4 is formed by combining a valley dynamic fluid model with a groundwater flow model and a surface water flow model; the groundwater flow model is formed by a first influence degree; the surface water flow model is formed by a second influence degree; The groundwater flow model uses the three-dimensional groundwater flow equation ,in is the permeability tensor, is the height of groundwater level change, is the water storage coefficient, is the source-sink term; The surface water flow model uses the Saint-Venant equations to describe the surface runoff process, and the formula is: ,here is the cross-sectional area of ​​water flow, It's traffic. is the distance along the river channel, is the net rain intensity.

8. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 6 is characterized in that: The hydrological-hydrodynamic simulation framework is formed by combining the valley dynamic fluid model with the groundwater flow model and the surface water flow model, and the formula is: ,in is the surface elevation, is the surface runoff flow, is the exchange rate, is the height of groundwater level change, and are the weight coefficients of the exchange rate between groundwater and surface water, respectively, which are used to quantify the interaction strength between the two.

9. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 1, characterized in that: The S5 comprises the following steps: S5.

1. Use the hydrological-hydrodynamic simulation framework to simulate rainfall scenarios of varying intensity and duration. S5.

2. Identify the surface runoff response parameters under different rainfall scenarios and extract the key factors affecting the surface runoff behavior; S5.

3. Combine valley micro-topography characteristic data to analyze the impact of valley topography on water flow paths and convergence points, and determine peak flood flow and frequency; S5.

4. Calculate the rate of change of flow per unit time through time series data under different rainfall scenarios, and identify mutation points and periodic fluctuations in flow changes.

10. The method for analyzing nonlinear runoff characteristics of a small watershed in a valley terrain according to claim 9, characterized in that: The peak flow and frequency in S5.3 are determined by comparing the simulation results with the changes in runoff flow under different rainfall events, identifying the conditions that lead to the formation of peaks, and using extreme value theory to calculate extreme peak flow and the probability of peak occurrence.

Images