A method for forecasting mountain flood disasters in small watersheds by coupling meteorology, hydrology and hydrodynamic

Abstract

The present invention discloses a method for forecasting mountain flood disasters in small watersheds with meteorological - hydrological - hydrodynamic coupling. This method constructs a multi - dimensional quantification index system for disaster - causing factors and disaster - bearing environments, combines a three - dimensional water system model and a smoothed particle hydrodynamics (SPH) hydrodynamic model to construct a disaster prediction ontology and extract characteristic data, optimizes the features and target variables through machine learning and deep learning, trains the extracted features, and constructs disaster scenarios based on historical mountain flood disaster data, thereby establishing a chain relationship among hydrometeorological data - hydrodynamic characteristics - water level and flow rate characterization - disaster situations. Through fine - scale hydrodynamic simulation, accurately considering the influence of sediment deposition and impurities, and combining with meteorological - hydrological models to provide detailed predictions of flood flow paths and speeds, the present invention realizes high - precision and real - time mountain flood risk prediction and the construction of multiple disaster scenarios, effectively improving the accuracy and response speed of mountain flood forecasting.

Description

Technical Field

[0001] The present invention relates to the field of hydrological prediction and hydrodynamic simulation, and particularly to a method for forecasting mountain flood disasters in small watersheds by coupling meteorology, hydrology, and hydrodynamics. Background Art

[0002] Mountain flood disasters are common and extremely destructive natural disasters in mountainous areas. They occur frequently, develop rapidly, and are difficult to predict, posing a serious threat to human life and property safety and social and economic development. Traditional mountain flood warning methods mainly rely on meteorological data (such as rainfall) and hydrological data (such as river water level, flow rate, etc.) for prediction, while considering the terrain and underlying surface factors of the watershed.

[0003] Statistical - based methods analyze historical meteorological and hydrological data and mountain flood occurrence records to establish statistical models for prediction. These methods are simple to operate and have a fast calculation speed. However, the prediction accuracy is limited by the representativeness of historical data, making it difficult to adapt to climate change and complex geographical environments, and unable to capture the complex non - linear relationships in the occurrence of mountain flood disasters. In recent years, machine learning methods have been widely applied to mountain flood warning. By learning a large amount of meteorological and hydrological data, complex non - linear relationships are established to improve prediction accuracy. To a certain extent, these methods overcome the deficiencies of traditional methods. However, their performance highly depends on the quality and quantity of training data, and they have poor interpretability of the model, making it difficult to provide an intuitive understanding of the physical process. Pure data - driven models may experience a significant decline in prediction ability when facing extreme meteorological conditions and unknown environmental changes, limiting their application effects in a dynamically changing environment. Hydrodynamic models simulate the hydraulic characteristics of river channels under different flow rates and water levels based on the principles of fluid mechanics, and can relatively accurately predict the flow path and inundation range of floods. However, traditional physical models cannot directly incorporate hydrometeorological elements such as rainfall, runoff generation and concentration, and soil permeability. There are limitations in the parameter calibration of complex river channel topography and riverbed characteristics during the initial data import and processing, increasing the difficulty and cost of practical applications. Summary of the Invention

[0004] To solve the above problems, the present invention proposes a method for forecasting mountain flood disasters in small watersheds by coupling meteorology, hydrology, and hydrodynamic forces. By combining the SPH hydrodynamic model and machine learning algorithms, a chain relationship of hydrometeorological data - hydrodynamic characteristics - water level and flow rate representation - disaster situation (flood inundation range, inundation depth, and mountain flood propagation speed) is constructed. This method organically combines meteorological data, hydrological data, and hydrodynamic characteristics, comprehensively considers various factors such as rainfall, hydrological monitoring data, terrain elevation, river cross-section, and riverbed characteristics, and combines hydrodynamic coupling mechanism modeling and machine learning algorithms to more comprehensively and accurately simulate the occurrence and development process of mountain flood disasters. Compared with ordinary machine learning algorithms, the meteorology-hydrology-hydrodynamic coupling method not only improves the prediction accuracy but also enhances the interpretability and robustness of the model, ensuring excellent prediction performance under complex river channel environments and changing meteorological conditions, and improving the reliability and practicality of the early warning system.

[0005] The present invention is realized through the following technical solutions: A method for forecasting mountain flood disasters in small watersheds by coupling meteorology, hydrology, and hydrodynamic forces, comprising the following steps:

[0006] A method for forecasting mountain flood disasters in small watersheds by coupling meteorology, hydrology, and hydrodynamic forces, comprising the following steps:

[0007] Step S1, comprehensively collect multi-source data and obtain three-dimensional spatial data through satellite images;

[0008] Step S2, clean the data to remove noise and missing values, and perform normalization and standardization processing to unify the data scale;

[0009] Step S3, select and quantify the main disaster-causing factors and disaster-forming environment factors, and comprehensively analyze and construct a multi-dimensional quantification index system;

[0010] Step S4, construct a three-dimensional water system model and a three-dimensional river channel hydrodynamic model, integrate terrain data, and segment the water system;

[0011] Step S5, use the smoothed particle hydrodynamics method SPH to construct a hydrodynamic model with segmented nodes; combine hydrological data and meteorological data to simulate the movement trajectories of floods under different scenarios;

[0012] Step S6, after running the SPH model, analyze the key indicators of the flood, extract and analyze the water flow characteristic indicators of each segment of the water system, select the corresponding characteristic data as secondary acquisition data, and integrate it into the input variables of the prediction model after data preprocessing;

[0013] Step S7, use the water flow characteristic indicators obtained from the SPH model as target variables, optimize the features and target variables, and train the extracted features to construct a chain relationship of hydrometeorological data - hydrodynamic characteristics - water level and flow rate representation - disaster situation;

[0014] Step S8: Validate the model and optimize the algorithm to obtain a trained comprehensive model;

[0015] Step S9: Deploy the trained comprehensive model to the real-time data processing system to forecast and issue early warning information in real time.

[0016] Furthermore, the step S1 includes the following sub-steps:

[0017] Step S1-1: Collect meteorological data, including but not limited to rainfall, rainfall intensity, wind speed, and wind direction, to obtain the meteorological conditions affecting the occurrence of mountain floods;

[0018] Step S1-2: Collect hydrological data, including but not limited to river water level, water flow velocity, and soil moisture, for analyzing the hydrological process and its impact on floods;

[0019] Step S1-3: Collect topographic data, including but not limited to topographic elevation, slope, and vegetation coverage rate, for analyzing the topographic features and vegetation conditions of the basin;

[0020] Step S1-4: Collect historical mountain flood disaster data, including but not limited to occurrence time, location, intensity, and disaster impact range, to establish the historical basis for the occurrence of disasters;

[0021] Step S1-5: Collect the basin water system map and riverbed map through satellite images to obtain river channel morphology, sediment distribution, and impurity information, providing spatial data support for the construction of the three-dimensional water system model.

[0022] Furthermore, the step S2 includes the following sub-steps:

[0023] Step S2-1: Clean the collected data to remove noise and missing values;

[0024] Step S2-2: Perform data normalization to unify the data scale;

[0025] Step S2-3: Smooth the time series data to eliminate short-term fluctuations;

[0026] Step S2-4: Standardize the riverbed morphology and sediment distribution data to ensure the quality of the input data for the hydrodynamic model.

[0027] Furthermore, the step S3 includes the following sub-steps:

[0028] Step S3-1: Select the main disaster-causing factors, including but not limited to rainfall, rainfall intensity, rainfall duration, temperature, soil moisture, and vegetation coverage, identify the relationship with the occurrence of mountain flood disasters through quantitative analysis; combine the historical meteorological data of different regions to fit the intensity, frequency, and spatial distribution models of rainfall;

[0029] Step S3-2: Select the disaster-forming environment factors within the basin, including but not limited to terrain elevation, slope, soil type, vegetation coverage rate, and underlying surface impact factors of the basin;

[0030] Step S3-3: Based on the multivariate analysis of meteorological factors and disaster-forming environment factors, construct a multi-dimensional quantitative index system.

[0031] Furthermore, the said Step S4 includes the following sub-steps:

[0032] Step S4-1: Utilize high-resolution satellite image data to extract the spatial information of the river system within the basin, obtain the precise morphology and distribution of the river channels, and generate a river system map of the basin;

[0033] Step S4-2: Generate a data base plate in tiff format based on the river system map of the basin, and import data information: digital terrain elevation, river channel cross-section distribution, river bed topographic map, river bank dike project data, sediment distribution, and impurity accumulation, to generate a three-dimensional model of the river system of the basin, comprehensively reflecting the three-dimensional structure of the river channels;

[0034] Step S4-3: Integrate the terrain data to ensure the accuracy and integrity of the three-dimensional model, and further refine the model for complex river bed areas;

[0035] Step S4-4: Divide the river system and define nodes, specifically: divide the river system of the basin into n segments, ensuring that each segment can reflect the actual terrain, basin structure, and hydrodynamic characteristics; specifically, the segmentation should be carried out according to the river channel length, basin slope, and terrain undulation to adapt to the water flow conditions in different regions; preferably, when segmenting the river system, carry out adaptive segmentation according to the hydrodynamic characteristics of different regions within the basin; set n nodes in each river system segment to describe key variables: changes in water flow, flow velocity, and water level; the nodes are set at key positions of the river channel, including but not limited to river channel bends and confluence points.

[0036] Furthermore, the said Step S5 includes the following sub-steps:

[0037] Step S5-1: Set boundary conditions and initial conditions for each segment of the river system respectively, specifically:

[0038] Step S5-1a: According to meteorological data and hydrological data, set appropriate boundary conditions for the starting and ending parts of each river system segment;

[0039] Step S5-1b: Set the initial water level state and flow velocity state according to historical data;

[0040] Step S5-1c: Initialize the particles of the segmented river channel model. Divide the three-dimensional water system model into several particles, where each particle represents a fluid unit in the water body, and assign physical properties to each particle: initial position, velocity, mass, and density. Represent the fluid by particles, and each particle carries physical quantities of the fluid: mass and velocity. The particles simulate the fluid motion through the interaction forces between them, thereby simulating the flood flow and propagation process. Ensure that the particle distribution can be refined to local areas.

[0041] Step S5-2: Select a kernel function. Select a kernel function suitable for flash flood simulation to define the interaction range between particles and determine the support radius of the kernel function to balance the calculation accuracy and efficiency.

[0042] Step S5-3: Perform density calculation. For each particle, calculate its local density using the following formula: ;

[0043] where, is the density of particle , is the mass of particle , is the kernel function, is the position of the i-th particle, is the position of the j-th particle, is the kernel radius;

[0044] Step S5-4: Perform pressure calculation. Calculate the pressure of each particle using the state equation with the following formula:

[0045] ;

[0046] where, is the pressure of particle , is the speed of sound, is the reference density;

[0047] Step S5-5: Perform force calculation. Calculate the pressure and the viscous force between particles using the following formulas:

[0048] ;

[0049] ;

[0050] where, is the viscosity coefficient, and are the velocities of particles and , is the pressure of the i-th particle. is the pressure of the j-th particle;

[0051] Step S5-6, perform external force and gravity calculations to calculate the external forces acting on the particles;

[0052] Step S5-7, perform total force and acceleration calculations, combine all forces, and calculate the total acceleration of the particles. The formula is as follows:

[0053] ;

[0054] where, is the gravitational force acting on the i-th particle; is the mass of the i-th particle;

[0055] Step S5-8, perform time integration, and use the time stepping method to update the velocity and position of the particles. The formula is as follows:

[0056] ;

[0057] ;

[0058] where, is the velocity of the i-th particle at time t, is the acceleration of the i-th particle at time t, is the time position, is the time step;

[0059] Step S5-9, perform boundary condition processing to prevent particles from crossing the solid boundary, and use the method of mirror particles or reflection force to handle boundary interactions.

[0060] Furthermore, the step S6 includes the following sub-steps:

[0061] Step S6-1, run the Smooth Particle Hydrodynamics (SPH) model to calculate the water flow trajectories under different rainfall intensities, durations, and terrain conditions; simulate the propagation process of flood waves, and through the model output, analyze the influence of the interaction between the propagation path of flood waves and the terrain and river channels on flood flow;

[0062] Step S6-2, based on the data output by the SPH model, calculate the propagation patterns of flash floods under different rainfall intensities, durations, and terrain conditions, and analyze the key indicators of floods: expansion speed, easily overflowing positions, river channel inundation areas; extract and analyze the water flow characteristics of each section of the water system, including key indicators: maximum water level, maximum flow rate, flood propagation speed, and predicted inundation area;

[0063] Step S6-3, use the key indicator parameters derived in step S6-2 as secondary acquisition data, perform noise reduction processing on the data, and perform normalization and standardization processing;

[0064] Step S6-4: Use a feature selection algorithm to screen the preprocessed secondary acquisition data and select the most predictive feature values.

[0065] Furthermore, the said Step S7 includes the following sub-steps:

[0066] Step S7-1: Based on statistical methods or machine learning algorithms, establish a non-linear relationship model between each disaster-causing factor and disaster-forming environment factor in the index system in Step S3-3 and the feature values in Step S6-4. Screen the factors most influential on the occurrence of mountain floods through the feature selection algorithm, quantify each factor, and calculate the sensitivity of each disaster-causing factor and disaster-forming environment factor.

[0067] Step S7-2: Use the inundation range, inundation depth, and propagation speed obtained from the SPH model as target variables, optimize the features and target variables through machine learning and deep learning, and train the extracted features to establish a multi-factor and multi-variable prediction model.

[0068] Step S7-3: Construct a chain relationship of hydrometeorological data - hydrodynamic characteristics - water level and flow characterization - disaster situation.

[0069] Step S7-4: Establish the occurrence mechanism of mountain flood disasters under different scenarios by analyzing the rainfall, flow changes, and disaster impacts in historical data.

[0070] Step S7-5: Use the generated disaster scenarios to simulate the mountain flood response under different rainfall intensities and durations and provide multiple disaster scenario predictions.

[0071] Furthermore, the said Step S8 includes the following sub-steps:

[0072] Step S8-1: Use the method of cross-validation to optimize the model parameters, prevent overfitting, and ensure the generalization ability of the model in different watershed environments.

[0073] Step S8-2: Improve the stability and accuracy of the model through the method of ensemble learning.

[0074] Step S8-3: Achieve synchronous prediction of multiple target variables to ensure the consistency and relevance of the prediction results.

[0075] Step S8-4: Use an independent test data set to verify the prediction performance of the model, and evaluate the accuracy and applicability of the model by comparing the prediction results with the actual observed data.

[0076] Step S8-5: Adjust the model structure and parameters according to the verification results and optimize the combination of hydrodynamic simulation and machine learning algorithms.

[0077] Further, the step S9 includes the following sub-steps:

[0078] Step S9-1: Deploy the trained comprehensive model to the real-time data processing system, and the model can receive and process the latest data input in real time;

[0079] Step S9-2: Receive real-time data, combine with the output of the hydrodynamic model, conduct mountain flood risk assessment and forecasting, and update the current flood dynamic prediction results in a timely manner;

[0080] Step S9-3: Combine with the SPH model to achieve three-dimensional visualization display of flood evolution, and convey early warning information through visualization technology with the simulation results.

[0081] The beneficial effects of the present invention are as follows:

[0082] Based on the combination of meteorological and hydrological information and the hydrodynamic model, the present invention continuously iterates and calibrates using historical disaster conditions and real-time monitoring data. On the one hand, it gives full play to the high-precision simulation and interpretability advantages of the physical model for the flood propagation process. On the other hand, it uses machine learning algorithms to deal with complex non-linear relationships and extreme conditions, forming a more robust and time-effective mountain flood prediction system; at the same time, by establishing a scenario-based mountain flood disaster scenario library, it provides accurate simulation predictions for the emergency department under different rainfall patterns, river channel conditions, and land use changes, significantly improving the accuracy, rapid response ability, and practicality of mountain flood forecasting in small mountainous basins. Description of the Drawings

[0083] Figure 1 is the overall flowchart of a method for forecasting mountain flood disasters in small basins based on meteorological-hydrological-hydrodynamic coupling of the present invention. Detailed Embodiments

[0084] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. In the following description and drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0085] The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are only examples of devices consistent with some aspects of the present invention as detailed in the appended claims. Each embodiment of this specification is described in a progressive manner.

[0086] As Figure 1 shown, the following technical solutions are proposed. According to a method for forecasting mountain flood disasters in small basins based on meteorological-hydrological-hydrodynamic coupling provided by the present application, it includes the following steps:

[0087] Step S1, comprehensively collect multi-source data, including obtaining meteorological data such as rainfall, rainfall intensity, temperature, humidity, etc., hydrological data such as river water level, water flow velocity, flow rate, etc., and obtaining three-dimensional spatial data such as watershed water systems, topographic elevation, and slope through satellite images;

[0088] Step S2, clean the data to remove noise and missing values, and perform normalization and standardization processing to unify the data scale to ensure the quality of the input data for the hydrodynamic model;

[0089] Step S3, select and quantify the main disaster-causing factors and disaster-forming environmental factors affecting mountain flood occurrence, and construct a multi-dimensional quantification index system based on the comprehensive analysis of these multiple factors and environmental elements;

[0090] Step S4, construct a three-dimensional water system model, establish a three-dimensional hydrodynamic model of the river channel based on digital topographic elevation, watershed water system map, river channel cross-section data, riverbed topographic map, and riverbank dike project data, integrate the topographic data, and segment the water system;

[0091] Step S5, use the smoothed particle hydrodynamics (SPH) method to construct a hydrodynamic model with segmented nodes to simulate the water flow movement, propagation, and overflow during the mountain flood process. Combine hydrological data (such as watershed water level, flow rate, etc.) and meteorological data (such as rainfall, intensity, etc.) to simulate the movement trajectories of floods under different scenarios, and reflect the generation and evolution of mountain floods;

[0092] Step S6, after running the SPH model, analyze key indicators such as the flood expansion speed, possible overflow locations, and river channel inundation areas, extract and analyze the water flow characteristics of each segment of the water system, such as key indicators such as the maximum water level, maximum flow rate, flood propagation speed, and predicted inundation area, so as to quantify the responses of disaster-causing factors and disaster-forming environments under different scenarios, select corresponding characteristic data as secondary acquisition data, and integrate it into the input variables of the prediction model after data preprocessing;

[0093] Step S7, use the inundation range, inundation depth, and propagation speed obtained from the SPH model as target variables, optimize the features and target variables through machine learning and deep learning, and train the extracted features to construct a chain relationship of hydrometeorological data - hydrodynamic characteristics - water level and flow rate characterization - disaster situation (inundation range, inundation depth, and mountain flood propagation speed);

[0094] Step S8, model verification and algorithm optimization to ensure the reliability and accuracy of the prediction model;

[0095] Step S9, real-time forecasting and release to realize the practical application of the model and provide early warning information.

[0096] Furthermore, the specific content of step S1 is as follows:

[0097] Step S1-1, collect meteorological data, including rainfall, rainfall intensity, wind speed, wind direction, etc., to obtain the meteorological conditions affecting the occurrence of mountain floods;

[0098] Step S1-2, collect hydrological data, including river water level, water flow velocity, soil moisture, etc., for analyzing the hydrological process and its impact on floods;

[0099] Step S1-3, collect topographic data, including terrain elevation, slope, vegetation coverage, etc., to understand the topographic features and vegetation conditions of the basin;

[0100] Step S1-4, collect historical mountain flood disaster data, including occurrence time, location, intensity, disaster impact scope, etc., to establish the historical basis for the occurrence of disasters;

[0101] Step S1-5, collect the river system map and riverbed map of the basin through satellite images, obtain information on river channel morphology, sediment distribution and impurities, and provide spatial data support for the construction of a three-dimensional river system model.

[0102] Furthermore, the specific steps of step S2 are as follows:

[0103] Step S2-1, clean the collected data, remove noise and missing values, and ensure the integrity and accuracy of the data;

[0104] Step S2-2, perform data normalization processing to unify the data scale and facilitate comparison and integration between different data sources;

[0105] Step S2-3, smooth the time series data to eliminate short-term fluctuations and improve the stability of the data;

[0106] Step S2-4, standardize the riverbed morphology and sediment distribution data to ensure the quality of the input data for the hydrodynamic model.

[0107] Furthermore, the specific steps of step S3 are as follows:

[0108] Step S3-1, select the main disaster-causing factors, such as rainfall, rainfall intensity, rainfall duration, temperature, soil moisture, vegetation coverage, etc., and identify the relationship with the occurrence of mountain flood disasters through quantitative analysis; combine the historical meteorological data of different regions to fit the intensity, frequency and spatial distribution models of rainfall;

[0109] Step S3-2, select the disaster-bearing environment factors within the basin, including terrain elevation, slope, soil type, vegetation coverage, etc., and consider the underlying surface impact factors of the basin;

[0110] Step S3-3, based on the multivariate analysis of meteorological factors and disaster-bearing environment factors, construct a multi-dimensional quantitative index system;

[0111] Further, the specific steps of step S4 are as follows:

[0112] Step S4-1: Using high-resolution satellite image data, extract the spatial information of the river basin water system, obtain the precise morphology and distribution of the river channels, and generate a river basin water system map.

[0113] Step S4-2: Generate a data baseplate in tiff format based on the river basin water system map, import data information such as digital terrain elevation, river channel cross-section distribution, riverbed topographic map, riverbank dike engineering data, sediment distribution, and impurity accumulation, and generate a three-dimensional model of the river basin water system to comprehensively reflect the three-dimensional structure of the river channels.

[0114] Step S4-3: Integrate the terrain data to ensure the accuracy and integrity of the three-dimensional model, and pay special attention to the details of complex riverbed areas to improve the refinement degree of the model.

[0115] Step S4-4: Water system division and node definition, specifically:

[0116] Divide the river basin water system into n paragraphs to ensure that each paragraph can reflect the actual terrain, river basin structure, and hydrodynamic characteristics; specifically, the segmentation should be carried out according to factors such as river channel length, river basin slope, and terrain undulation to adapt to the water flow conditions in different regions; preferably, when segmenting the water system, adaptive segmentation can be carried out according to the hydrodynamic characteristics of different regions within the river basin to improve the calculation accuracy and stability of the model.

[0117] Set n nodes in each water system segment to describe key variables such as water flow changes, flow velocity, and water level. The nodes can be set at key positions of the river channel, such as river channel bends and confluence outlets.

[0118] Further, the specific steps of step S5 are as follows:

[0119] Step S5-1: Set boundary conditions and initial conditions for each water system segment of the water system respectively, specifically:

[0120] Step S5-1a: According to meteorological data (rainfall, intensity) and hydrological data (water level, flow rate, etc.), set appropriate boundary conditions for the starting and ending parts of each water system segment.

[0121] Step S5-1b: Set initial water level, flow velocity, and other states according to historical data.

[0122] Step S5-1c: Initialize the particles of the segmented river channel model. Divide the three-dimensional water system model into several particles, where each particle represents a fluid unit in the water body, and assign physical properties such as initial position, velocity, mass, and density to each particle. Represent the fluid by particles, with each particle carrying physical quantities such as the mass and velocity of the fluid. The particles simulate the fluid motion through mutual forces, thereby simulating the flood flow and propagation process. Ensure that the distribution of particles can be refined to local areas (such as river bends and complex terrain areas) to improve the adaptability of the model under different basin conditions.

[0123] Step S5-2: Select a kernel function. Select a kernel function suitable for flash flood simulation, such as the cubic B-spline kernel function, to define the interaction range between particles and determine the support radius of the kernel function to balance the calculation accuracy and efficiency.

[0124] Step S5-3: Perform density calculation. For each particle, calculate its local density using the following formula:

[0125]

[0126] where, is the density of particle , is the mass of particle , is the kernel function, is the position of the i-th particle, is the position of the j-th particle, is the kernel radius;

[0127] Step S5-4: Perform pressure calculation. Calculate the pressure of each particle using the state equation with the following formula:

[0128]

[0129] where, is the pressure of particle , is the speed of sound, is the reference density;

[0130] Step S5-5: Perform force calculation. Calculate the pressure and viscous force between particles using the following formula:

[0131]

[0132]

[0133] where, is the viscosity coefficient, and is the velocity of the particle and is the velocity of the is the pressure of the i-th particle, is the pressure of the j-th particle;

[0134] Step S5-6, perform external force and gravity calculations to calculate the external forces acting on the particles;

[0135] Step S5-7, perform total force and acceleration calculations, combine all forces, and calculate the total acceleration of the particles. The formula is as follows:

[0136] ;

[0137] is the gravitational force acting on the i-th particle; is the mass of the i-th particle;

[0138] Step S5-8, perform time integration, and use a time stepping method (such as Leapfrog or Verlet integration) to update the velocity and position of the particles. The formula is as follows:

[0139] ;

[0140] ;

[0141] where, is the velocity of the i-th particle at time t, is the acceleration of the i-th particle at time t, is the time position, is the time step;

[0142] Step S5-9, perform boundary condition processing to implement boundary conditions such as riverbanks and levees, prevent particles from crossing solid boundaries, and use the method of mirror particles or reflection forces to handle boundary interactions;

[0143] Furthermore, perform SPH modeling for each section of the water system separately. The outlet boundary of the previous river section serves as the inlet boundary of the next river section, and the nodes at the confluence positions of tributaries are modeled separately to ensure the continuity and accuracy of the hydrodynamic model.

[0144] Furthermore, in Step S6, by simulating the propagation and dynamic changes of floods, the response characteristics of hazard-causing factors and disaster-forming environments under different scenarios are quantified. Specifically:

[0145] Step S6-1: Run the Smoothed Particle Hydrodynamics (SPH) model to calculate the water flow trajectories under different rainfall intensities, durations, and terrain conditions; simulate the propagation process of flood waves, considering the coupled effects of precipitation, basin water level, terrain, and soil moisture, etc.; through the model output, analyze the influence of the interaction between the propagation path of flood waves and the terrain and river channels on flood flow.

[0146] Step S6-2: Based on the data output by the SPH model, calculate the propagation patterns of flash floods under different rainfall intensities, durations, and terrain conditions, and analyze key indicators such as the flood expansion speed, possible overflow locations, and river channel inundation areas; extract and analyze the water flow characteristics of each section of the water system, such as key indicators like the maximum water level, maximum flow rate, flood propagation speed, and predicted inundation area; in addition, in addition to the flood propagation speed and inundation depth, factors such as sediment movement, reflection, and refraction of flood waves can also be considered for modeling.

[0147] Step S6-3: Take the key index parameters derived in Step S6-2 as secondary acquisition data, perform noise reduction processing on the data, and perform normalization and standardization processing to ensure that these parameters can be effectively utilized in the subsequent feature extraction and selection steps.

[0148] Step S6-4: Use feature selection algorithms (such as principal component analysis, genetic algorithms) to screen the preprocessed secondary acquisition data, and select the most predictive features, such as the spatio-temporal distribution of extreme rainfall events, the river water level and flow rate change curve, the riverbed evolution curve, etc.

[0149] Furthermore, Step S7 is specifically as follows:

[0150] Step S7-1: Based on statistical methods or machine learning algorithms (such as regression analysis), establish a non-linear relationship model between each hazard-causing factor and disaster-forming environment factor in the index system in Step S3-3 and the eigenvalue in Step S6-4. Screen the factors that have the most influence on the occurrence of flash floods through the feature selection algorithm, quantify each factor, and calculate the sensitivity of each hazard-causing factor and disaster-forming environment factor.

[0151] Step S7-2: Take the inundation range, inundation depth, and propagation speed obtained from the SPH model as target variables, optimize the features and target variables through machine learning and deep learning (such as heuristic search, evolutionary strategy, and Bayesian optimization), and train the extracted features to establish a multi-factor and multi-variable prediction model.

[0152] Preferably, multiple machine learning algorithms can be used for training respectively, for comparison, and the model most suitable for the modeling basin can be selected.

[0153] Step S7-3, construct the chain relationship of hydrometeorological data - hydrodynamic characteristics - water level and flow rate characterization - disaster situation (flooded area, inundation depth, and flash flood propagation speed);

[0154] Step S7-4, establish the flash flood disaster occurrence mechanism under different scenarios by analyzing the rainfall, flow rate changes, and disaster impacts in historical data;

[0155] Step S7-5, use the generated disaster scenarios to simulate the flash flood response under different rainfall intensities and durations, and provide predictions of multiple possible disaster scenarios.

[0156] Furthermore, the specific steps of Step S8 are as follows:

[0157] Step S8-1, optimize the model parameters using methods such as cross-validation to prevent overfitting, ensure the generalization ability of the model in different watershed environments, and improve the stability and reliability of the model;

[0158] Step S8-2, further improve the stability and accuracy of the model through ensemble learning methods such as Bagging and Boosting;

[0159] Preferably, use Bagging to train multiple base learners through multiple sampling with replacement, and average or vote on their prediction results to reduce the variance of the model;

[0160] Preferably, use Boosting to train the base learners step by step, focusing on the mispredictions of the previous learner to reduce the bias of the model;

[0161] Furthermore, using these two methods simultaneously can make full use of the advantages of multiple base learners, reduce both the variance and bias of the model, thereby improving the generalization ability and robustness of the model, and ensuring its excellent flash flood disaster prediction performance under different watershed river channel environments and changing meteorological conditions.

[0162] Step S8-3, achieve synchronous prediction of multiple target variables, ensure the consistency and relevance of the prediction results, and improve the adaptability of the model to complex flood situations.

[0163] Step S8-4, use an independent test dataset to verify the prediction performance of the model, especially its performance in complex river channel environments. By comparing the prediction results with actual observed data, evaluate the accuracy and applicability of the model.

[0164] Step S8-5, adjust the model structure and parameters according to the verification results, optimize the combination of hydrodynamic simulation and machine learning algorithms, improve the overall prediction ability of the integrated model, and ensure its effectiveness in practical applications.

[0165] Furthermore, the specific steps of Step S9 are as follows:

[0166] Step S9-1: Deploy the trained comprehensive model into the real-time data processing system to ensure that the model can receive and process the latest data inputs in real time;

[0167] Step S9-2: Receive real-time data, combine with the output of the hydrodynamic model, conduct mountain flood risk assessment and forecasting, and update the prediction results in a timely manner to reflect the current flood dynamics;

[0168] Step S9-3: Combine with the SPH model to achieve three-dimensional visualization of flood evolution, and visually present the simulation results through visualization technology to help relevant departments and the public better understand and respond to mountain flood disasters;

[0169] Step S9-4: Release the forecast information through various channels, inform relevant departments and the public in a timely manner, and ensure the accurate transmission of early warning information.

[0170] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method, characterized in that: The steps include: Step S1, comprehensively collect multi-source data and obtain three-dimensional spatial data of watershed water system, terrain elevation and slope through satellite images; Step S2, cleaning the data to remove noise and missing values, and normalizing and standardizing the data to unify the data scale; Step S3, selecting and quantifying the main disaster-causing factors and disaster-pregnant environmental factors that affect the occurrence of flash floods, and constructing a multidimensional quantitative indicator system based on a comprehensive analysis of multiple factors and environmental elements; Step S4, constructing a three-dimensional water system model, establishing a three-dimensional water dynamic model of the river based on digital terrain elevation, watershed water system map, river section data, riverbed topographic map, and riverbank embankment engineering data, integrating terrain data, and dividing the water system into sections; Step S5, using the smooth particle method SPH to construct a segmented and noded hydrodynamic model to simulate the water flow movement, propagation and overflow during the flash flood process; combining hydrological data and meteorological data to simulate the movement trajectory of floods under different scenarios; Step S6, after running the SPH model, analyzing the key indicators of the flood and extracting and analyzing the water flow characteristic indicators of each section of the water system, specifically, based on the data output by the SPH model, calculating the propagation mode of the flash flood under different rainfall intensities, durations and terrain conditions, and analyzing the key indicators of the flood: expansion speed, overflow locations, and river flooding areas; Extract and analyze the flow characteristics of each section of the river system, including key indicators: maximum water level, maximum flow, flood propagation speed, and expected inundation area; thereby quantifying the response of disaster-causing factors and disaster-prone environments under different scenarios, and using the extracted key indicators as secondary collection data, which are integrated into the prediction model input variables after data preprocessing; Step S7 uses the water flow characteristic index obtained from the SPH model as the target variable, optimizes the target variable through machine learning and deep learning, and trains the extracted features to build a chain relationship of hydrological and meteorological data-hydrodynamic characteristics-water level and flow representation-disaster situation; specifically, it includes the following sub-steps: Step S7-1, based on statistical methods or machine learning algorithms, a nonlinear relationship model between each disaster-causing factor and disaster-pregnant environmental factor in the multidimensional quantitative index system and the key indicator with the most predictive ability is established, the factors that have the greatest impact on the occurrence of flash floods are screened through a feature selection algorithm, and each factor is quantified to calculate the sensitivity of each disaster-causing factor and disaster-pregnant environmental factor; Step S7-2: the maximum water level, maximum flow, flood propagation speed, and expected flooded area obtained from the SPH model are used as target variables, the target variables are optimized through machine learning and deep learning, and the extracted features are trained to establish a multi-factor and multi-variable prediction model; Step S7-3, constructing a chain relationship of hydrological and meteorological data - hydrodynamic characteristics - water level and flow representation - disaster situation; Step S7-4, by analyzing the rainfall, flow changes and disaster impacts in historical data, establish the flash flood disaster occurrence mechanism under different scenarios; Step S7-5, using the generated disaster scenario, simulating flash flood responses under different rainfall intensities and durations, and providing multiple disaster scenario predictions; Step S8, verifying the model and optimizing the algorithm to obtain a trained comprehensive model; Step S9, deploy the trained comprehensive model to the real-time data processing system to forecast and issue warning information in real time.

2. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S1 includes the following sub-steps: Step S1-1, collecting meteorological data, including but not limited to rainfall, rainfall intensity, wind speed, and wind direction, to obtain meteorological conditions that affect the occurrence of flash floods; Step S1-2, collecting hydrological data, including but not limited to river water level, water flow velocity, and soil moisture, for analyzing hydrological processes and their impact on floods; Step S1-3, collecting terrain data, including but not limited to terrain elevation, slope, and vegetation coverage, for analyzing the terrain characteristics and vegetation conditions of the watershed; Step S1-4, collecting historical flash flood disaster data, including but not limited to the time, location, intensity, and scope of disaster impact, to establish a historical basis for the occurrence of disasters; Step S1-5, collecting watershed water system maps and riverbed maps through satellite images, obtaining river channel morphology, sediment distribution and impurity information, and providing spatial data support for the construction of a three-dimensional water system model.

3. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S2 includes the following sub-steps: Step S2-1, cleaning the collected data to remove noise and missing values; Step S2-2, performing data normalization processing to unify the data scale; Step S2-3, smoothing the time series data to eliminate short-term fluctuations; Step S2-4, standardize the riverbed morphology and sediment distribution data to ensure the input data quality of the hydrodynamic model.

4. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S3 includes the following sub-steps: Step S3-1, selecting major disaster-causing factors, including but not limited to rainfall, rainfall intensity, rainfall duration, temperature, soil moisture and vegetation cover, and identifying their relationship with flash flood disasters through quantitative analysis; combining historical meteorological data from different regions to fit the intensity, frequency and spatial distribution model of rainfall; Step S3-2, selecting disaster-prone environmental factors in the watershed, including but not limited to terrain elevation, slope, soil type, vegetation coverage, and underlying surface influencing factors of the watershed; Step S3-3, construct a multidimensional quantitative indicator system based on the multivariate analysis of the disaster-causing factors and disaster-pregnant environmental factors in the indicator system.

5. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S4 includes the following sub-steps: Step S4-1, using high-resolution satellite image data, extracting spatial information of the watershed system, obtaining the precise shape and distribution of the river channel, and generating a watershed system map; Step S4-2, based on the watershed system map, a data base in tiff format is generated, and data information is imported: digital terrain elevation, river section distribution, riverbed topography, riverbank embankment engineering data, sediment distribution and impurity accumulation, to generate a three-dimensional model of the watershed system, which fully reflects the three-dimensional structure of the river; Step S4-3, integrating terrain data to ensure the accuracy and completeness of the 3D model. The model needs to be further refined for complex riverbed areas; Step S4-4, water system division and node definition, specifically: divide the watershed water system into n sections, ensuring that each section can reflect the actual topography, watershed structure and hydrodynamic characteristics; specifically, it should be segmented according to the length of the river channel, the slope of the watershed, and the undulating terrain to adapt to the water flow conditions in different areas; when segmenting the water system, adaptive segmentation should be performed according to the hydrodynamic characteristics of different areas in the watershed; n nodes are set in each water system segment to describe key variables: changes in water flow, flow velocity and water level; nodes are set at key positions of the river channel, including but not limited to river channel bends and confluences.

6. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S5 includes the following sub-steps: Step S5-1, setting boundary conditions and initial conditions for each section of the water system, specifically: Step S5-1a, according to the meteorological data and hydrological data, set appropriate boundary conditions for the start and end of each water system segment; Step S5-1b, setting the initial water level state and flow rate state according to historical data; Step S5-1c, perform particle initialization on the segmented river model, divide the three-dimensional water system model into several particles, each particle represents a fluid unit in the water body, and assign physical properties to each particle: initial position, velocity, mass and density; particles are used to represent fluids, and each particle carries the physical quantities of the fluid: mass and velocity. The particles simulate the fluid movement through interaction forces, thereby simulating the flow and propagation process of floods; ensure that the distribution of particles can be refined to local areas; Step S5-2, selecting a kernel function, selecting a kernel function suitable for flash flood simulation, used to define the interaction range between particles, and determining the support radius of the kernel function to balance calculation accuracy and efficiency; Step S5-3, perform density calculation, for each particle, calculate its local density, the formula is as follows: ρ i =∑ j m j W(|r i -r j |,h); Among them, ρ i is the density of particle i, m j is the mass of particle j, W is the kernel function, r i is the position of the ith particle, r j is the position of the jth particle, h is the nuclear radius; Step S5-4, perform pressure calculation, and use the state equation to calculate the pressure of each particle. The formula is as follows: P i =c 2 (r i -p0); Among them, P i is the pressure of particle i, c is the speed of sound, and ρ0 is the reference density; Step S5-5, perform force calculation to calculate the pressure between particles and viscosity The formula is as follows: Where μ is the viscosity coefficient, v j and v i is the velocity of particles i and j, P i is the pressure of the ith particle, P j is the pressure of the jth particle; Step S5-6, performing external force and gravity calculation to calculate the external force on the particle; Step S5-7, calculate the total force and acceleration, combine all forces, and calculate the total acceleration of the particle. The formula is as follows: in, is the gravitational force acting on the i-th particle; m i is the mass of the ith particle; Step S5-8, perform time integration, and use the time stepping method to update the velocity and position of the particle. The formula is as follows: v i (t+Δt)=v i (t)+a i (t)Δt; r i (t+Δt)=r i (t)+v i (t+Δt)Δt; Among them, v i is the velocity of the ith particle at time t, a i (t) is the acceleration of the ith particle at time t, t is the time position, and Δt is the time step; Step S5-9, performing boundary condition processing to prevent particles from crossing solid boundaries, and using mirror particles or reflection force methods to process boundary interactions.

7. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S6 includes the following sub-steps: Step S6-1, running the smooth particle method SPH model to calculate the water flow trajectory under different rainfall intensities, durations and terrain conditions; simulating the propagation process of flood waves, and analyzing the impact of the interaction between the propagation path of flood waves and the terrain and river channels on flood flow through the model output; Step S6-2, based on the data output by the SPH model, calculate the propagation mode of flash floods under different rainfall intensities, durations and terrain conditions, and analyze the key indicators of floods: expansion speed, overflow locations, and river flooding areas; Extract and analyze the flow characteristics of each section of the river system, including key indicators: maximum water level, maximum flow, flood propagation speed, and expected flooded area; Step S6-3, using the key indicator parameters derived in step S6-2 as secondary collection data, performing noise reduction, normalization and standardization on the data; Step S6-4, using a feature selection algorithm to screen the pre-processed secondary collection data and select the key indicators with the greatest predictive power.

8. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S8 comprises the following sub-steps: Step S8-1, using the cross-validation method to optimize the model parameters, prevent overfitting, and ensure the generalization ability of the model in different watershed environments; Step S8-2, improving the stability and accuracy of the model through an integrated learning method; Step S8-3, achieving synchronous prediction of multiple target variables to ensure consistency and relevance of prediction results; Step S8-4, using an independent test data set to verify the prediction performance of the model, and evaluating the accuracy and applicability of the model by comparing the prediction results with the actual observation data; Step S8-5, adjust the model structure and parameters according to the verification results to optimize the combination of hydrodynamic simulation and machine learning algorithm.

9. The meteorological-hydrological-hydrodynamic coupling small watershed flash flood disaster forecasting method according to claim 1 is characterized in that: The step S9 includes the following sub-steps: Step S9-1, deploying the trained comprehensive model to a real-time data processing system, wherein the model can receive and process the latest data input in real time; Step S9-2, receiving real-time data, combining the output of the hydrodynamic model, conducting flash flood risk assessment and forecasting, and timely updating the current flood dynamic results; Step S9-3, combined with the SPH model, realizes the three-dimensional visualization of flood evolution, and conveys the warning information through the simulation results through visualization technology.

Images