A method and system for basin rainfall runoff forecasting based on neural network

Abstract

The present invention relates to the field of rainfall runoff simulation technology, and discloses a basin rainfall runoff forecasting method and system based on neural network, the method comprising obtaining terrain basic data, measured rainfall runoff process diagram and historical rainfall runoff data; inputting the historical rainfall runoff data into a back propagation neural network to obtain initial model parameters of a rainfall runoff model; obtaining optimal model parameters according to the initial model parameters by using an optimal precision algorithm; using the optimal model parameters as preset parameters of a rainfall runoff model to obtain an optimal rainfall runoff model; and calculating rainfall runoff data according to the terrain basic data and the measured rainfall runoff process diagram, using the optimal rainfall runoff model and a preset runoff transmission model. The method improves the accuracy of the predicted runoff results by combining an artificial neural network to correct the parameters of the hydrological model.

Description

Technical Field

[0001] The present invention relates to the technical field of rainfall runoff simulation, and in particular to a basin rainfall runoff forecasting method and system based on a neural network. Background Art

[0002] Hydrological models are systems used to generate various computational results. They can simulate complex hydrological processes and, in turn, generate simulated runoff for a watershed. To meet application requirements, the model must fully interpret the correlation between the inputs and the results of its internal system. Hydrological models can be categorized by their development history into empirical, conceptual, and distributed models. Empirical models, with their relatively few input factors, are easier to master, leading to their long history and widespread application. However, they lack a physical foundation and are limited in their ability to accurately model hydrological processes. Conceptual models consider hydrological processes from multiple perspectives, including physics, chemistry, biology, ecology, and fluid mechanics, establishing input-output relationships. However, due to their relatively simple input-output relationships, they can only be applied to watersheds with simple hydrological processes and strong regularity. A series of distributed hydrological models have been developed to address complex watersheds. These models divide the watershed into grids, assume that the watershed response function is the same for each grid, and focus their simulations on analyzing and predicting the spatial variation of input factors. Distributed hydrological models offer higher simulation accuracy but are limited by the complexity of their input factors. Precipitation is the primary input factor for the model. Precipitation information is primarily obtained through regional rainfall maps regularly released by the Meteorological Bureau using satellite remote sensing and radar. However, due to the insufficient spatial resolution of this data, it is difficult to use for accurate rainfall forecasts, resulting in reduced basin-wide simulation accuracy. With the development of data mining technology, artificial intelligence (AI) is gradually being integrated with hydrological models. Training models with large datasets improves their forecast accuracy.

[0003] The application of data mining technology in the fields of hydrology and water resources has been extensively studied, providing guidance for the development of water resource allocation, hydrological forecasting, and other fields. Currently, the integration of artificial intelligence (AI) and hydrological models primarily involves applying AI methods, such as artificial neural networks (ANNs), support vector machines (SVMs), and genetic algorithms (GAs), to the parameter calibration and input factor selection of traditional models to improve model simulation accuracy. However, this integration primarily relies on traditional conceptual hydrological models for analysis, resulting in limited parameter information and an inability to fully reflect dynamic spatial changes, leading to inaccurate predictions. Summary of the Invention

[0004] The present invention provides a basin rainfall runoff forecasting method, system, electronic equipment and storage medium based on a neural network, so as to improve the prediction accuracy of the hydrological model.

[0005] In a first aspect, in order to solve the above technical problems, the present invention provides a basin rainfall runoff forecasting method based on a neural network, comprising:

[0006] Obtain basic terrain data, measured rainfall and runoff process maps, and historical rainfall and runoff data;

[0007] Inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of a rainfall-runoff model;

[0008] According to the initial model parameters, the optimal model parameters are obtained by using the best precision algorithm;

[0009] Using the optimal model parameters as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model;

[0010] According to the topographic basic data and the measured rainfall runoff process diagram, rainfall runoff data is calculated using the optimal rainfall runoff model and a preset runoff transmission model.

[0011] In an optional embodiment, inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of the rainfall-runoff model includes:

[0012] The historical rainfall runoff data include historical rain gauge rainfall data and historical outlet measured flow;

[0013] By means of parameter sensitivity analysis, the parameters of the rainfall-runoff model are selected, and then the initial values ​​of these parameters are randomly set as the initial weights and thresholds of the back propagation neural network;

[0014] Inputting the historical rain gauge rainfall data and the historical outlet measured flow into a rainfall runoff model to perform rainfall runoff simulation;

[0015] Calculating the error value of the back propagation neural network according to the simulation result;

[0016] The back propagation neural network performs weight and threshold training through a gradient descent method, completes parameter optimization, and obtains initial model parameters of the rainfall runoff model.

[0017] In an optional embodiment, obtaining optimal model parameters by using an optimal precision algorithm based on the initial model parameters includes:

[0018] ;

[0019] in, is the relative error of runoff, is the runoff volume at the i-th moment calculated by the model, is the measured runoff at the i-th moment, N is the total amount of runoff data;

[0020] ;

[0021] in, is the Nasher efficiency coefficient, is the average value of the measured runoff at the i-th moment, is the runoff volume at the i-th moment calculated by the model, is the measured runoff at the i-th moment;

[0022] Using the runoff relative error and the Nasher efficiency coefficient as rainfall fitting accuracy indicators of the optimal accuracy algorithm;

[0023] If the rainfall fitting accuracy index does not meet the preset threshold, the initial model parameters are input into the reverse prediction neural network for iteration, and the optimal model parameters are re-output.

[0024] In an optional embodiment, the step of using the optimal model parameters as preset parameters of a rainfall-runoff model to obtain the optimal rainfall-runoff model includes:

[0025] Calculating a confluence relationship matrix between hypercubes based on the optimal model parameters, and determining an optimal rainfall-runoff model based on the confluence relationship matrix;

[0026] The optimal rainfall-runoff model is a confluence relationship matrix. If there are n hypercubes, the confluence relationship matrix C is a Matrix, where It represents the proportion of water flow contribution of the j-th hypercube to the i-th hypercube.

[0027] In an optional embodiment, the calculating and obtaining rainfall runoff data based on the terrain basic data and the measured rainfall runoff process map using an optimal rainfall runoff model and a preset runoff transmission model includes:

[0028] Dividing the watershed into a number of hypercubes according to the basic terrain data, and calculating the terrain factor of each hypercube;

[0029] Calculating the water storage capacity of the hypercube based on the terrain factors;

[0030] According to the measured rainfall runoff process diagram, the rainfall is divided into a number of hypercubes, and the inflow of each cube is calculated;

[0031] If the water storage capacity is less than the inflow, then calculating the outflow of each hypercube according to the terrain factor and the inflow;

[0032] ;

[0033] in, is the outflow of the i-th unit at the j-th moment, is the drainage area of ​​the ith unit, is the inflow of the i-th unit at the j-th moment;

[0034] inputting the outflow into an optimal rainfall-runoff model to output a hypercube flow;

[0035] Inputting the hypercube flow into a runoff transport model to output rainfall runoff data;

[0036] The runoff transmission model is used to simulate the flow of water in a region, and the rainfall runoff data is obtained by solving the linear tank storage equation and the water balance equation.

[0037] In an optional embodiment, before dividing the rainfall into a plurality of hypercubes according to the measured rainfall-runoff process diagram and calculating the inflow of each cube, the method further includes:

[0038] A series of distributed rain gauges are set at the center point of the top surface of each hypercube;

[0039] ;

[0040] in, is the flow rate, is the rainfall amount in the ith region into flow, is the outlet interception flow of the ith area, and n is the number of rain gauges set.

[0041] In an optional embodiment, the terrain factors include slope, shape factor, water flow path length, and water flow path slope.

[0042] In a second aspect, the present invention provides a basin rainfall runoff forecasting system based on a neural network, comprising:

[0043] Data acquisition module, used to obtain basic terrain data, measured rainfall runoff process map and historical rainfall runoff data;

[0044] An initial parameter module, used for inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of the rainfall-runoff model;

[0045] A parameter optimization module is used to obtain the optimal model parameters according to the initial model parameters through an optimal precision algorithm;

[0046] A model optimization module is used to use the optimal model parameters as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model;

[0047] The result output module is used to calculate the rainfall runoff data according to the terrain basic data and the measured rainfall runoff process map, using the optimal rainfall runoff model and the preset runoff transmission model.

[0048] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any one of the above-mentioned neural network-based basin rainfall runoff forecasting methods.

[0049] In a fourth aspect, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned neural network-based basin rainfall runoff forecasting methods.

[0050] Compared to existing technologies, the present invention has the following advantages: The present invention discloses a neural network-based method and system for basin rainfall-runoff forecasting. The method comprises obtaining basic terrain data, a measured rainfall-runoff process map, and historical rainfall-runoff data; inputting the historical rainfall-runoff data into a back-propagation neural network to obtain initial model parameters for a rainfall-runoff model; using the initial model parameters, an optimal model parameter is obtained using an optimal precision algorithm; the optimal model parameter is used as the preset parameters for the rainfall-runoff model to obtain an optimal rainfall-runoff model; and based on the basic terrain data and the measured rainfall-runoff process map, rainfall-runoff data is calculated using the optimal rainfall-runoff model and a preset runoff transport model. By constructing a rainfall-runoff model based on the coupling of DRIVE and RAPID, combined with hypercube terrain factors, parameters, and surface rainfall data, accurate simulation of flow in each sub-region within the basin is achieved. This method effectively captures the process by which rainfall events are converted into surface runoff and accurately calculates water flow transfer between hypercubes using a confluence relationship matrix, thereby improving the accuracy of flood forecasting and water resource management. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of a method for forecasting rainfall and runoff in a watershed based on a neural network provided by the first embodiment of the present invention;

[0052] Figure 2This is a schematic structural diagram of a neural network-based watershed rainfall runoff forecasting system provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] Hydrological models are systems used to generate various computational results. They can simulate complex hydrological processes and, in turn, generate simulated runoff for a watershed. To meet application requirements, the model must fully interpret the correlation between the inputs and the results of its internal system. Hydrological models can be categorized by their development history into empirical, conceptual, and distributed models. Empirical models, with their relatively few input factors, are easier to master, leading to their long history and widespread application. However, they lack a physical foundation and are limited in their ability to accurately model hydrological processes. Conceptual models consider hydrological processes from multiple perspectives, including physics, chemistry, biology, ecology, and fluid mechanics, establishing input-output relationships. However, due to their relatively simple input-output relationships, they can only be applied to watersheds with simple hydrological processes and strong regularity. A series of distributed hydrological models have been developed to address complex watersheds. These models divide the watershed into grids, assume that the watershed response function is the same for each grid, and focus their simulations on analyzing and predicting the spatial variation of input factors. Distributed hydrological models offer higher simulation accuracy but are limited by the complexity of their input factors. Precipitation is the primary input factor for the model. Precipitation information is primarily obtained through regional rainfall maps regularly released by the Meteorological Bureau using satellite remote sensing and radar. However, due to the insufficient spatial resolution of this data, it is difficult to use for accurate rainfall forecasts, resulting in reduced basin-wide simulation accuracy. With the development of data mining technology, artificial intelligence (AI) is gradually being integrated with hydrological models. Training models with large datasets improves their forecast accuracy.

[0055] The application of data mining technology in the fields of hydrology and water resources has been extensively studied, providing guidance for the development of water resource allocation, hydrological forecasting, and other fields. Currently, the integration of artificial intelligence (AI) and hydrological models primarily involves applying AI methods, such as artificial neural networks (ANNs), support vector machines (SVMs), and genetic algorithms (GAs), to the parameter calibration and input factor selection of traditional models to improve model simulation accuracy. However, this integration primarily relies on traditional conceptual hydrological models for analysis, resulting in limited parameter information and an inability to fully reflect dynamic spatial changes, leading to inaccurate predictions.

[0056] Reference Figure 1 The first embodiment of the present invention provides a method for forecasting rainfall and runoff in a watershed based on a neural network, comprising the following steps:

[0057] S11, obtaining basic terrain data, measured rainfall runoff process map and historical rainfall runoff data;

[0058] S12, inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of a rainfall-runoff model;

[0059] S13, obtaining optimal model parameters according to the initial model parameters through an optimal precision algorithm;

[0060] S14, using the optimal model parameters as preset parameters of a rainfall-runoff model to obtain an optimal rainfall-runoff model;

[0061] S15, calculating and obtaining rainfall runoff data based on the topographic basic data and the measured rainfall runoff process map using the optimal rainfall runoff model and a preset runoff transmission model.

[0062] In step S11, basic terrain data, measured rainfall runoff process map and historical rainfall runoff data are obtained.

[0063] In one embodiment, the basic terrain data includes a digital elevation model, a topographic map and vector data, as well as remote sensing data. The digital elevation model can be obtained from the official website of ASTER GDEM (Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model), which is a public source. GIS software can be used to download and process these data. The topographic map and vector data are downloaded and processed by GIS software in the National Administration of Surveying, Mapping and Geoinformation, local surveying and mapping departments or geographic information service centers. Remote sensing data comes from satellite images. The measured rainfall runoff process map is recorded by hydrological stations operated by national or local water conservancy departments and meteorological bureaus, including rainfall, flow and other data. Historical rainfall runoff data are recorded in the China Meteorological Administration or local meteorological stations for a long period of time. The required historical rainfall runoff data can also be found through the global data set provided by the World Meteorological Organization.

[0064] It's worth noting that a Digital Elevation Model (DEM) is a key dataset used in Geographic Information Systems (GIS) to represent Earth's surface topography. A DEM represents ground elevation information through an ordered array of numerical values, with each value representing the elevation of a specific location. This data can be used to generate topographic maps, conduct terrain analysis, simulate water flow paths, and assess land use. It's typically stored using a regular rectangular grid, but grid points are not characteristic points and cannot accurately represent microtopographic features. Alternatively, an irregular triangulated network (TMN) can be used to more accurately represent complex terrain and avoid loss of interpolation accuracy.

[0065] In step S12, the historical rainfall-runoff data is input into a back propagation neural network to obtain initial model parameters of the rainfall-runoff model.

[0066] In one embodiment, the historical rainfall runoff data is input into a back propagation neural network to obtain initial model parameters of the rainfall runoff model, including: the historical rainfall runoff data includes historical rain gauge rainfall data and historical outlet measured flow; through parameter sensitivity analysis, the parameters of the rainfall runoff model are selected, and then the initial values ​​of these parameters are randomly set as the initial weights and thresholds of the back propagation neural network; the historical rain gauge rainfall data and the historical outlet measured flow are input into the rainfall runoff model to perform rainfall runoff simulation; the error value of the back propagation neural network is calculated based on the simulation results; the back propagation neural network performs weight and threshold training through the gradient descent method to complete parameter optimization and obtain the initial model parameters of the rainfall runoff model.

[0067] In one implementation, network weights and biases are randomly initialized; input data is passed through the network, and the output of each node is calculated. A loss function is used to calculate the difference between the predicted value and the actual value; the gradient of the loss with respect to each weight and bias is calculated using the chain rule; the weights and biases are updated to minimize the loss function; and the forward and backward propagation processes are repeated until a predetermined number of training rounds is reached or convergence conditions are met. Rectified LinearUnit is selected as the activation function, mean squared error is used as the loss function, and stochastic gradient descent is used as the optimization algorithm. Model performance is evaluated on the validation set. If the model performs poorly on the validation set, consider adjusting the network structure, hyperparameters, or performing regularization.

[0068] In step S13, the optimal model parameters are obtained according to the initial model parameters through the precision optimization algorithm.

[0069] In one embodiment, obtaining optimal model parameters using an optimal precision algorithm based on the initial model parameters includes:

[0070] ;

[0071] in, is the relative error of runoff, is the runoff volume at the i-th moment calculated by the model, is the measured runoff at the i-th moment, N is the total amount of runoff data;

[0072] ;

[0073] in, is the Nasher efficiency coefficient, is the average value of the measured runoff at the i-th moment, is the runoff volume at the i-th moment calculated by the model, is the measured runoff at the i-th moment;

[0074] Using the runoff relative error and the Nasher efficiency coefficient as rainfall fitting accuracy indicators of the optimal accuracy algorithm;

[0075] If the rainfall fitting accuracy index does not meet the preset threshold, the initial model parameters are input into the reverse prediction neural network for iteration, and the optimal model parameters are re-output.

[0076] It's worth noting that the runoff relative error is a simple measure of the difference between model simulation results and observed data; smaller values ​​indicate smaller absolute deviations between model predictions and measured data. The Nash-Sutcliffe efficiency coefficient is a more complex evaluation metric that considers not only the average deviation between model predictions and observed data, but also their variance relative to the mean of the observed data. The Nash-Sutcliffe efficiency coefficient typically ranges from negative infinity to 1. A Nash-Sutcliffe efficiency coefficient close to 1 indicates a high model fit. A Nash-Sutcliffe efficiency coefficient of 0 indicates that the model performs no better than using the mean of the observed data as a prediction. A Nash-Sutcliffe efficiency coefficient less than 0 indicates that the model's performance is even worse than using a simple mean prediction. The runoff relative error and the Nash-Sutcliffe efficiency coefficient serve as loss functions in the backpropagation neural network. The calculated accuracy metrics are compared against a preset threshold. If the accuracy metric meets the requirements, the current parameters are considered the optimal solution.

[0077] In step S14, the optimal model parameters are used as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model.

[0078] It is worth noting that the rainfall-runoff model described uses the DRIVE model, a distributed rainfall-runoff model primarily designed to simulate rainfall-runoff processes within a watershed. It incorporates multiple factors, such as topography, soil type, and land use, to provide detailed flow path analysis and runoff generation predictions. The DRIVE model is designed to improve the accuracy of flood forecasting and support water resources management and environmental impact assessments. It first divides the watershed into multiple subregions (hypercubes), each with its own unique topography, soil, and land use characteristics. This allows the model to more accurately capture the impact of spatial variability on runoff generation. The model considers multiple hydrological processes, including rainfall infiltration, surface runoff, and subsurface runoff, realistically simulating actual hydrological processes. The model utilizes the Green-Ammert method to calculate runoff generation. Based on topographic data, flow paths are determined, a confluence relationship matrix is ​​constructed, and flow transfer between subregions is calculated, ultimately yielding the discharge at the watershed outlet. The model is calibrated using historical observations, and parameters are adjusted to improve simulation accuracy. The model's performance is evaluated using a validation dataset to ensure its reliability under diverse conditions. In the present invention, it is used to calculate the total flow.

[0079] It is worth noting that Darcy's law is the basic equation that describes the flow of fluids in porous media. This law shows that the seepage velocity of water in saturated soil is proportional to the hydraulic gradient. In groundwater dynamics, Darcy's law is used to calculate the flow of groundwater in aquifers. For example, when simulating the impact of different soil types on groundwater recharge in a basin, Darcy's law can be used to estimate the groundwater flow velocity in different areas. In the rainfall-runoff process, Darcy's law can be used to simulate the soil infiltration process. When the rainfall exceeds the infiltration capacity of the soil, the excess water will form surface runoff. The permeability of the soil directly affects the rate of generation of surface runoff. As the basic principle of the infiltration velocity of water in the DRIVE model, it is used in the present invention to calculate the total flow rate.

[0080] In step S15, rainfall runoff data is calculated based on the terrain basic data and the measured rainfall runoff process map using the optimal rainfall runoff model and the preset runoff transmission model.

[0081] In one embodiment, the calculating of rainfall runoff data based on the terrain basic data and the measured rainfall runoff process map using an optimal rainfall runoff model and a preset runoff transmission model includes:

[0082] Based on the basic terrain data, the watershed is divided into a plurality of hypercubes, and the terrain factor of each hypercube is calculated; the water storage capacity of the hypercube is calculated based on the terrain factor; based on the measured rainfall-runoff process map, the rainfall is divided into a plurality of hypercubes, and the inflow of each cube is calculated; if the water storage capacity is less than the inflow, the outflow of each hypercube is calculated based on the terrain factor and the inflow;

[0083] ;

[0084] in, is the outflow of the i-th unit at the j-th moment, is the drainage area of ​​the ith unit, is the inflow of the i-th unit at the j-th moment;

[0085] Inputting the outflow into an optimal rainfall-runoff model to output a hypercube flow; inputting the hypercube flow into a runoff transport model to output rainfall-runoff data;

[0086] The runoff transmission model is used to simulate the flow of water in a region, and the rainfall runoff data is obtained by solving the linear tank storage equation and the water balance equation.

[0087] It is worth noting that the runoff transport model uses the rapid model. This is a model for parallel calculation of runoff transport, which is designed to simulate the flow process of surface water within a watershed. This model is particularly suitable for runoff simulation in large-scale watersheds and long time series, and can efficiently process large amounts of data. Its core is a river confluence algorithm based on the Muskingum method, which is a method widely used in hydrology to simulate the movement of water in rivers. Furthermore, the core of the Muskingum method is to obtain the finite difference formula of the flow by jointly solving the linear channel storage equation and the water balance equation. The runoff transport model can obtain rainfall runoff data by inputting the hypercube flow output. The parameters required by the runoff transport model can be optimized using the same optimization method as the optimal model parameters of the rainfall runoff data, or the particle swarm optimization algorithm can be used for parameter optimization.

[0088] It is worth noting that, before dividing the rainfall into a plurality of hypercubes according to the measured rainfall runoff process diagram and calculating the inflow of each cube, the method further includes:

[0089] A series of distributed rain gauges are set at the center point of the top surface of each hypercube;

[0090] ;

[0091] in, is the flow rate, is the rainfall amount in the ith region into flow, is the outlet interception flow of the ith area, and n is the number of rain gauges set.

[0092] It is worth noting that rainfall is usually highly variable in space, especially in areas with complex terrain or changeable climate. A single rain gauge cannot accurately reflect the actual rainfall distribution in the entire basin or region. Hydrological models are very sensitive to the quality of input data. Data collected using distributed rain gauges can provide more refined input for the model, thereby improving the accuracy of the rainfall-runoff model output. In the present invention, the use of rain gauges as the data source for cube inflow calculation has higher reliability than the data source obtained from a single rain gauge. At the same time, rain gauge data can be combined with other hydrological observation data to gain a more comprehensive understanding of the water cycle process. This helps to establish more complex hydrological models and improve the understanding and prediction capabilities of hydrological phenomena. At the same time, for areas without data, rain gauges help establish basic rainfall data records. This helps to understand the climate characteristics, rainfall patterns and seasonal changes in the area. In areas without data, data obtained by installing rain gauges can also be used to preliminarily calibrate model parameters to improve the applicability of the model in the local area.

[0093] It is worth noting that the present invention improves the computational efficiency of hydrological models by using hypercube cells instead of traditional Thiessen polygons. The core advantage of this approach is that it simplifies spatial data processing and analysis while effectively capturing topographic features. Each hypercube cell can be viewed as a small square of uniform size, covering the entire study area. Because each cell is of regular shape, the relationships between adjacent cells are clear and consistent, enabling algorithmic optimizations such as parallel processing and vectorization. For large-scale watershed simulations, this structured grid system can better utilize modern computer hardware resources, such as multi-core processors and graphics processing units. Despite the simple use of hypercube cells, the invention does not ignore the importance of topographic factors. Key topographic factors such as slope, shape factor, flow path length, and flow path slope can still be accurately calculated using high-resolution digital elevation model data and assigned to each hypercube cell. This approach preserves topographic details even within a regular grid structure. When processing rainfall data, observations from distributed rain gauges can be assigned to corresponding hypercube cells. This not only improves the spatial resolution of rainfall data but also makes rainfall field interpolation more intuitive and efficient.

[0094] It is worth noting that the terrain factors mentioned include slope, shape factor, flow path length, and flow path slope. These factors influence the formation, convergence, and transmission of water flows. Slope refers to the degree of ground inclination, usually expressed as a percentage or angle. The greater the slope, the faster the surface water flow, reducing the opportunity for water infiltration and increasing runoff. Shape factor generally refers to the morphological characteristics of the watershed, such as the aspect ratio and circularity of the watershed. The shape of the watershed can affect the concentration and speed of water flow. For example, a long and narrow watershed causes water to converge more quickly at the outlet, while a circular or nearly circular watershed results in a more even distribution of water flow. Abstracting watershed morphological characteristics into shape factors facilitates subsequent calculations. Flow path length: Flow path length is the shortest surface distance from a point to the watershed outlet. A longer flow path means water must travel a longer distance to reach the watershed outlet, resulting in more water infiltration and evaporation losses, thereby reducing the final runoff. The flow path slope is the average slope along the flow path. Similar to slope, the slope of a flow path affects flow velocity and energy, which in turn influences erosion and runoff formation. Steeper path slopes result in faster flow velocities and greater erosive forces. Topographic factors are used to define hydrological response units within a sub-basin to improve the accuracy of rainfall-runoff model predictions.

[0095] In summary, this invention proposes a method for improving hydrological model parameter optimization, aiming to enhance the accuracy and reliability of runoff predictions. The core of this method is to optimize model parameters using high-precision rainfall data and topographic factors to better simulate surface runoff processes. First, the invention utilizes a distributed network of rain gauges to collect detailed rainfall information. This distributed monitoring system can capture the spatial variability of rainfall and provide more accurate input data, which is particularly important in areas with complex terrain or variable climates. Second, the invention considers multiple topographic factors, including slope, shape factor, flow path length, and flow path slope. These topographic factors significantly influence the formation, convergence, and transport of water flows. By incorporating them into the model, it can more realistically reflect actual hydrological processes. To evaluate model performance, the invention introduces two key metrics: runoff relative error (RE) and Nash-Sutcliffe efficiency (NSE). These metrics measure the difference between model predictions and measured runoff in terms of absolute and relative error, respectively. When the initial model parameters fail to meet the preset accuracy requirements, the invention employs an iterative optimization method based on a backpropagation neural network to adjust the parameters. This method utilizes the powerful nonlinear fitting ability of neural networks, and through continuous learning and adjustment, gradually approaches the optimal solution. The entire optimization process includes initializing parameters, running the hydrological model, calculating accuracy indicators, checking whether the threshold conditions are met, and if not, updating the parameters through the neural network until satisfactory accuracy or the maximum number of iterations is achieved. In addition, the invention also emphasizes the importance of installing rain gauges in areas without data. In these areas where historical data records are lacking, rain gauges not only provide basic data support, but also provide valuable real-time information for flood forecasting, water resources management, scientific research, etc. In this way, even in the case of scarce data, an effective hydrological model can be established, providing a scientific basis for local flood prevention and disaster reduction, agricultural irrigation, urban planning and other aspects. In general, this invention improves the prediction accuracy of rainfall runoff models by comprehensively applying advanced monitoring technology, complex terrain factor analysis and intelligent optimization algorithms.

[0096] Reference Figure 2 The second embodiment of the present invention provides a basin rainfall runoff forecasting system based on a neural network, comprising:

[0097] Data acquisition module, used to obtain basic terrain data, measured rainfall runoff process map and historical rainfall runoff data;

[0098] An initial parameter module, used for inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of the rainfall-runoff model;

[0099] A parameter optimization module is used to obtain the optimal model parameters according to the initial model parameters through an optimal precision algorithm;

[0100] A model optimization module is used to use the optimal model parameters as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model;

[0101] The result output module is used to calculate the rainfall runoff data according to the terrain basic data and the measured rainfall runoff process map, using the optimal rainfall runoff model and the preset runoff transmission model.

[0102] Preferably, the data acquisition module is used to:

[0103] Obtain basic terrain data, measured rainfall and runoff process maps, and historical rainfall and runoff data.

[0104] Preferably, the initial parameter module is used to:

[0105] The historical rainfall-runoff data is input into the back propagation neural network to obtain the initial model parameters of the rainfall-runoff model, including:

[0106] The historical rainfall runoff data include historical rain gauge rainfall data and historical outlet measured flow;

[0107] By means of parameter sensitivity analysis, the parameters of the rainfall-runoff model are selected, and then the initial values ​​of these parameters are randomly set as the initial weights and thresholds of the back propagation neural network;

[0108] Inputting the historical rain gauge rainfall data and the historical outlet measured flow into a rainfall runoff model to perform rainfall runoff simulation;

[0109] Calculating the error value of the back propagation neural network according to the simulation result;

[0110] The back propagation neural network performs weight and threshold training through a gradient descent method, completes parameter optimization, and obtains initial model parameters of the rainfall runoff model.

[0111] Preferably, the parameter optimization module is used to:

[0112] According to the initial model parameters, the optimal model parameters are obtained by using the best precision algorithm, including:

[0113] ;

[0114] in, is the relative error of runoff, is the runoff volume at the i-th moment calculated by the model, is the measured runoff at the i-th moment, N is the total amount of runoff data;

[0115] ;

[0116] in, is the Nasher efficiency coefficient, is the average value of the measured runoff at the i-th moment, is the runoff volume at the i-th moment calculated by the model, is the measured runoff at the i-th moment;

[0117] Using the runoff relative error and the Nasher efficiency coefficient as rainfall fitting accuracy indicators of the optimal accuracy algorithm;

[0118] If the rainfall fitting accuracy index does not meet the preset threshold, the initial model parameters are input into the reverse prediction neural network for iteration, and the optimal model parameters are re-output.

[0119] Preferably, the model optimization module is used to:

[0120] The optimal model parameters are used as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model, including:

[0121] Calculating a confluence relationship matrix between hypercubes based on the optimal model parameters, and determining an optimal rainfall-runoff model based on the confluence relationship matrix;

[0122] The optimal rainfall-runoff model is a confluence relationship matrix. If there are n hypercubes, the confluence relationship matrix C is a Matrix, where It represents the proportion of water flow contribution of the j-th hypercube to the i-th hypercube.

[0123] Preferably, the result output module is used to:

[0124] According to the terrain basic data and the measured rainfall runoff process map, the optimal rainfall runoff model and the runoff transmission model are used to calculate rainfall runoff data, including:

[0125] Dividing the watershed into a number of hypercubes according to the basic terrain data, and calculating the terrain factor of each hypercube;

[0126] Calculating the water storage capacity of the hypercube based on the terrain factors;

[0127] According to the measured rainfall runoff process diagram, the rainfall is divided into a number of hypercubes, and the inflow of each cube is calculated;

[0128] If the water storage capacity is less than the inflow, then calculating the outflow of each hypercube according to the terrain factor and the inflow;

[0129] ;

[0130] in, is the outflow of the i-th unit at the j-th moment, is the drainage area of ​​the ith unit, is the inflow of the i-th unit at the j-th moment;

[0131] inputting the outflow into an optimal rainfall-runoff model to output a hypercube flow;

[0132] Inputting the hypercube flow into a runoff transport model to output rainfall runoff data;

[0133] The runoff transmission model is used to simulate the flow of water in a region, and the rainfall runoff data is obtained by solving the linear tank storage equation and the water balance equation.

[0134] Preferably, before dividing the rainfall into a plurality of hypercubes according to the measured rainfall runoff process diagram and calculating the inflow of each cube, the method further comprises:

[0135] A series of distributed rain gauges are set at the center point of the top surface of each hypercube;

[0136] ;

[0137] in, is the flow rate, is the rainfall amount in the ith region into flow, is the outlet interception flow of the ith area, and n is the number of rain gauges set.

[0138] Preferably, the terrain factors include slope, shape factor, water flow path length, and water flow path slope.

[0139] It should be noted that the neural network-based watershed rainfall and runoff forecasting system provided in an embodiment of the present invention is used to execute all the process steps of the neural network-based watershed rainfall and runoff forecasting method in the above embodiment. The working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0140] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a result output program. When the processor executes the computer program, the steps in the above-mentioned embodiments of the neural network-based watershed rainfall runoff forecasting method are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the data acquisition module.

[0141] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0142] The electronic device may be a computing device such as a desktop computer, notebook, PDA, or smart tablet. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those described above, or a combination of certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, and the like.

[0143] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of the electronic device and connects various parts of the entire electronic device using various interfaces and lines.

[0144] The memory can be used to store the computer programs and / or modules. The processor implements the various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and accessing the data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0145] If the module / unit integrated into the electronic device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0146] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0147] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A basin rainfall runoff forecasting method based on neural network, characterized in that: include: Obtain basic terrain data, measured rainfall-runoff process maps and historical rainfall-runoff data; Inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of a rainfall-runoff model; According to the initial model parameters, the optimal model parameters are obtained by using the best accuracy algorithm; Using the optimal model parameters as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model; According to the terrain basic data and the measured rainfall runoff process diagram, the rainfall runoff data is calculated using the optimal rainfall runoff model and the preset runoff transmission model; Wherein, the optimal model parameters are used as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model, including: According to the optimal model parameters, a confluence relationship matrix between hypercubes is calculated, and an optimal rainfall-runoff model is determined according to the confluence relationship matrix; The optimal rainfall-runoff model is a confluence relationship matrix. If there are n hypercubes, the confluence relationship matrix C is a Matrix, where It represents the proportion of water flow contribution of the jth hypercube to the i-th hypercube; The method of calculating the rainfall runoff data based on the terrain basic data and the measured rainfall runoff process diagram using the optimal rainfall runoff model and the preset runoff transmission model includes: According to the basic terrain data, the watershed is divided into a number of hypercubes, and the terrain factor of each hypercube is calculated; Calculating the water storage capacity of the hypercube based on the terrain factors; According to the measured rainfall runoff process diagram, the rainfall is divided into a number of hypercubes, and the inflow of each cube is calculated; If the water storage capacity is less than the inflow, the outflow of each hypercube is calculated according to the terrain factor and the inflow; in, is the outflow of the ith unit at the jth moment, is the drainage area of ​​the ith unit, is the inflow of the ith unit at the jth moment; inputting the outflow into an optimal rainfall-runoff model to output a hypercube flow; inputting the hypercube flow into a runoff transport model to output rainfall runoff data; The runoff transport model is used to simulate the flow of water in a region, and the rainfall runoff data is obtained by solving the linear trough storage equation and the water balance equation.

2. The method for basin rainfall runoff forecasting based on neural network according to claim 1 is characterized in that: The historical rainfall-runoff data is input into the back propagation neural network to obtain the initial model parameters of the rainfall-runoff model, including: The historical rainfall runoff data include historical rain gauge rainfall data and historical outlet measured flow; By means of parameter sensitivity analysis, the parameters of the rainfall-runoff model are selected, and then the initial values ​​of these parameters are randomly set as the initial weights and thresholds of the back propagation neural network; Inputting the historical rain gauge rainfall data and the historical outlet measured flow into a rainfall runoff model to perform rainfall runoff simulation; Calculate the error value of the back propagation neural network according to the simulation result; The back propagation neural network performs weight and threshold training through the gradient descent method, completes parameter optimization, and obtains initial model parameters of the rainfall runoff model.

3. The method for basin rainfall runoff forecasting based on neural network according to claim 1 is characterized in that: The step of obtaining the optimal model parameters by using the best accuracy algorithm according to the initial model parameters includes: in, is the relative error of runoff, is the runoff at the i-th moment calculated by the model, is the measured runoff at the i-th moment, N is the total amount of runoff data; in, is the Nasher efficiency coefficient, is the average value of the measured runoff at the i-th moment, is the runoff at the i-th moment calculated by the model, is the measured runoff at the i-th moment; The runoff relative error and the Nasher efficiency coefficient are used as rainfall fitting accuracy indicators of the optimal accuracy algorithm; If the rainfall fitting accuracy index does not meet the preset threshold, the initial model parameters are input into the reverse prediction neural network for iteration, and the optimal model parameters are re-output.

4. The method for basin rainfall runoff forecasting based on neural network according to claim 1 is characterized in that: Before dividing the rainfall into a plurality of hypercubes according to the measured rainfall runoff process diagram and calculating the inflow of each cube, the method further comprises: A series of distributed rain gauges are set at the center point of the top surface of each hypercube; in, For flow rate, is the rainfall in the ith area into the flow, is the outlet interception flow of the ith area, and n is the number of rain gauges set.

5. The method for basin rainfall runoff forecasting based on neural network according to claim 1 is characterized in that: The terrain factors include slope, shape factor, water flow path length, and water flow path slope.

6. A basin rainfall runoff forecasting system based on neural network, characterized in that: include: Data acquisition module, used to obtain basic terrain data, measured rainfall runoff process diagrams and historical rainfall runoff data; An initial parameter module, used for inputting the historical rainfall-runoff data into a back propagation neural network to obtain initial model parameters of a rainfall-runoff model; A parameter optimization module, used to obtain optimal model parameters according to the initial model parameters through an optimal accuracy algorithm; A model optimization module, used for taking the optimal model parameters as preset parameters of a rainfall-runoff model to obtain an optimal rainfall-runoff model; A result output module is used to calculate rainfall runoff data according to the terrain basic data and the measured rainfall runoff process map, using the optimal rainfall runoff model and the preset runoff transmission model; Wherein, the optimal model parameters are used as preset parameters of the rainfall-runoff model to obtain the optimal rainfall-runoff model, including: According to the optimal model parameters, a confluence relationship matrix between hypercubes is calculated, and an optimal rainfall-runoff model is determined according to the confluence relationship matrix; The optimal rainfall-runoff model is a confluence relationship matrix. If there are n hypercubes, the confluence relationship matrix C is a Matrix, where It represents the proportion of water flow contribution of the jth hypercube to the i-th hypercube; The method of calculating the rainfall runoff data based on the terrain basic data and the measured rainfall runoff process diagram using the optimal rainfall runoff model and the preset runoff transmission model includes: According to the basic terrain data, the watershed is divided into a number of hypercubes, and the terrain factor of each hypercube is calculated; Calculating the water storage capacity of the hypercube based on the terrain factors; According to the measured rainfall runoff process diagram, the rainfall is divided into a number of hypercubes, and the inflow of each cube is calculated; If the water storage capacity is less than the inflow, the outflow of each hypercube is calculated according to the terrain factor and the inflow; in, is the outflow of the ith unit at the jth moment, is the drainage area of ​​the ith unit, is the inflow of the ith unit at the jth moment; inputting the outflow into an optimal rainfall-runoff model to output a hypercube flow; inputting the hypercube flow into a runoff transport model to output rainfall runoff data; The runoff transport model is used to simulate the flow of water in a region, and the rainfall runoff data is obtained by solving the linear trough storage equation and the water balance equation.

7. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the neural network-based basin rainfall runoff forecasting method as described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the basin rainfall runoff forecasting method based on neural network as described in any one of claims 1 to 5.

Images