Two-dimensional flood risk prediction method and device, electronic equipment, medium and product

Abstract

The invention relates to a two-dimensional flood risk prediction method and device, electronic equipment, a medium and a product. The method comprises the following steps: acquiring multi-modal data in a target drainage basin; obtaining a simulation runoff result based on a plurality of pre-established hydrological models according to the multi-modal data in the target drainage basin, and establishing a hydrological-machine learning coupling model based on the simulation runoff result; the method comprises the following steps: simulating a basin hydrological process in a future scene through a hydrological-machine learning coupling model to obtain a basin hydrological process simulation result in the future scene, and extracting flood duration and flood volume characteristic values in a plurality of warming horizontal time windows by utilizing a preset optimal model based on the basin hydrological process simulation result; and obtaining a two-dimensional flood risk prediction result based on the flood duration and the flood volume characteristic values under the plurality of warming horizontal time windows, thereby solving the problems of low flood risk prediction accuracy and poor applicability in related technologies, and improving the flood risk prediction accuracy and applicability.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrological disaster assessment, and in particular to a two-dimensional flood risk prediction method, device, electronic equipment, medium and product. Background Art

[0002] As global climate change increasingly impacts regional ecological environments and socioeconomics, global warming, a key characteristic of climate change, alters the thermal and dynamical environment of the climate system, impacts the global hydrological cycle, and increases extreme precipitation events and flooding. In regions most severely affected by floods, the rate of warming is far higher than the global average, with temperatures expected to rise by 4°C by the end of this century. This poses a serious threat to flood control, water supply security, food security, energy security, and ecological and environmental security. A deeper understanding of flood evolution and its socioeconomic impacts under a warming climate is crucial for future extreme climate disaster risk prediction, disaster prevention and mitigation, and adaptation management.

[0003] In related technologies, the evolution of future floods is studied by combining a global climate model ensemble and a river basin hydrological model. The near-term (2021-2040), medium-term (2041-2060) and long-term (2081-2100) are usually used to represent future climate projections.

[0004] However, due to the global warming effect caused by increased emissions of greenhouse gases such as carbon dioxide, the frequency and intensity of floods are exhibiting new characteristics, challenging the applicability and accuracy of traditional methods for flood risk prediction in a warming climate. Furthermore, relevant technologies do not consider flood risk assessment under specific warming levels and generally focus on single attributes such as flood peak, flood volume, or flood duration, failing to reflect the multivariate characteristics of floods. This issue urgently needs to be addressed. Summary of the Invention

[0005] The present invention provides a two-dimensional flood risk prediction method, device, electronic device, medium and product to solve the problems of low flood risk prediction accuracy and poor applicability in related technologies, and improve the accuracy and applicability of flood risk prediction.

[0006] The first aspect of the present invention provides a two-dimensional flood risk prediction method, comprising the following steps: obtaining multimodal data within a target watershed; obtaining simulated runoff results based on a plurality of pre-established hydrological models according to the multimodal data within the target watershed, and establishing a hydrological-machine learning coupling model based on the simulated runoff results; simulating the watershed hydrological process under future scenarios through the hydrological-machine learning coupling model to obtain simulation results of the watershed hydrological process under future scenarios, and based on the watershed hydrological process simulation results, extracting flood duration and flood volume characteristic values ​​under multiple warming level time windows using a preset optimal model, and obtaining a two-dimensional flood risk prediction result based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows.

[0007] Furthermore, in some embodiments, the multimodal data includes multiple global climate model simulation data, meteorological data of the ERA5 data set, socioeconomic data and daily flow series data of the target basin control hydrological station in the target basin. The simulated runoff results are obtained based on the multimodal data in the target basin based on multiple pre-established hydrological models, and a hydrological-machine learning coupling model is established based on the simulated runoff results, including: calculating relative humidity and specific humidity results based on the meteorological data of the ERA5 data set; and obtaining the simulated runoff results through multiple preset hydrological models based on the daily flow series data of the target basin control hydrological station and the meteorological data of the ERA5 data set; determining the simulated runoff duration of the machine learning model based on the simulated runoff results and the pre-collected daily measured runoff results according to a preset correlation threshold; and correcting the simulated runoff results based on a long short-term memory neural network with a coupled attention mechanism to obtain the hydrological-machine learning coupling model.

[0008] Furthermore, in some embodiments, before simulating the basin hydrological process under the future scenario through the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under the future scenario, it also includes: based on the decoding-encoding deviation correction method, correcting the multiple global climate model simulation data to obtain a corrected meteorological simulation series results.

[0009] Furthermore, in some embodiments, the simulation of the watershed hydrological process under the future scenario by the hydrological-machine learning coupling model to obtain the simulation results of the watershed hydrological process under the future scenario includes: based on the corrected meteorological simulation series results, simulating the watershed hydrological process under the future scenario by the hydrological-machine learning coupling model to obtain a series of daily runoff results under the future climate change scenario; based on the daily runoff series results under the future climate change scenario, obtaining the simulation results of the watershed hydrological process under the future scenario.

[0010] Furthermore, in some embodiments, before extracting flood duration and flood volume characteristic values ​​under multiple warming level time windows based on the simulation results of the basin hydrological process using a preset optimal model, it also includes: calculating the time windows under different warming levels based on the temperature data in the corrected meteorological simulation series results; fusing the corrected meteorological simulation series results with the machine learning model to obtain a fused machine learning model; determining the model parameters of the fused machine learning model based on a preset adaptive model selection mechanism, and obtaining the preset optimal model according to the determination results.

[0011] Furthermore, in some embodiments, the two-dimensional flood risk prediction result is obtained based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows, including: constructing a marginal distribution function of flood duration and flood volume based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows, and constructing a Copula joint distribution function of flood duration and flood volume under inconsistency conditions; calculating the joint recurrence period of flood duration and flood volume under different warming level time windows based on the marginal distribution function of flood duration and flood volume and the Copula joint distribution function of flood duration and flood volume under inconsistency conditions, to obtain the two-dimensional flood risk prediction result.

[0012] According to the two-dimensional flood risk prediction method provided by the embodiment of the present invention, basic meteorological and hydrological data of the watershed are first collected, and then humidity-related data are derived, the hydrological model is calibrated and a machine learning model is constructed; then, a meteorological simulation series under the climate change scenario is obtained through decoding-encoding bias correction, and the model is driven to simulate future hydrological processes; flood characteristic values ​​are extracted based on the optimal model, and a joint probability distribution function is constructed; finally, a two-dimensional flood risk prediction result is obtained, which solves the problems of low flood risk prediction accuracy and poor applicability in related technologies and improves the accuracy and applicability of flood risk prediction.

[0013] The second aspect of the present invention provides a two-dimensional flood risk prediction device, wherein the device includes: an acquisition module for acquiring multimodal data within a target watershed; a mathematical modeling module for obtaining simulated runoff results based on a plurality of pre-established hydrological models according to the multimodal data within the target watershed, and establishing a hydrological-machine learning coupling model based on the simulated runoff results; a prediction module for simulating the watershed hydrological process under future scenarios through the hydrological-machine learning coupling model to obtain simulation results of the watershed hydrological process under future scenarios, and based on the watershed hydrological process simulation results, using a preset optimal model to extract flood duration and flood volume characteristic values ​​under multiple warming level time windows, and obtaining a two-dimensional flood risk prediction result based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows.

[0014] Furthermore, in some embodiments, the mathematical modeling module is specifically used to: calculate relative humidity and specific humidity results based on the meteorological data of the ERA5 dataset; and obtain the simulated runoff results through multiple preset hydrological models based on the daily flow series data of the target watershed control hydrological station and the meteorological data of the ERA5 dataset; determine the simulated runoff duration of the machine learning model based on the simulated runoff results and the pre-collected daily measured runoff results according to a preset correlation threshold; and correct the simulated runoff results based on the long short-term memory neural network with a coupled attention mechanism to obtain the hydrological-machine learning coupling model.

[0015] Furthermore, in some embodiments, before simulating the basin hydrological process under the future scenario through the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under the future scenario, the prediction module is also used to: based on the decoding-encoding deviation correction method, correct the multiple global climate model simulation data to obtain a corrected meteorological simulation series results.

[0016] Furthermore, in some embodiments, the prediction module is also used to: based on the corrected meteorological simulation series results, simulate the basin hydrological process under the future scenario through the hydrological-machine learning coupling model to obtain a series of daily runoff results under the future climate change scenario; based on the daily runoff series results under the future climate change scenario, obtain the basin hydrological process simulation results under the future scenario.

[0017] Furthermore, in some embodiments, before extracting flood duration and flood volume characteristic values ​​under multiple warming level time windows based on the simulation results of the basin hydrological process using a preset optimal model, the prediction module is also used to: calculate the time windows under different warming levels based on the temperature data in the corrected meteorological simulation series results; fuse the corrected meteorological simulation series results with the machine learning model to obtain a fused machine learning model; determine the model parameters of the fused machine learning model based on a preset adaptive model selection mechanism, and obtain the preset optimal model according to the determination results.

[0018] Furthermore, in some embodiments, the prediction module is further used to: construct a marginal distribution function of flood duration and flood volume based on the characteristic values ​​of flood duration and flood volume under the multiple warming level time windows, and construct a Copula joint distribution function of flood duration and flood volume under inconsistency conditions; calculate the joint recurrence period of flood duration and flood volume under different warming level time windows based on the marginal distribution function of flood duration and flood volume and the Copula joint distribution function of flood duration and flood volume under inconsistency conditions, and obtain the two-dimensional flood risk prediction result

[0019] According to the two-dimensional flood risk prediction device provided by the embodiment of the present invention, basic meteorological and hydrological data of the watershed are first collected, and then humidity-related data are derived, the hydrological model is calibrated and a machine learning model is constructed; then, a meteorological simulation series under a climate change scenario is obtained through decoding-encoding bias correction, and the model is driven to simulate future hydrological processes; flood characteristic values ​​are extracted based on the optimal model, and a joint probability distribution function is constructed; finally, a two-dimensional flood risk prediction result is obtained, which solves the problems of low flood risk prediction accuracy and poor applicability in related technologies and improves the accuracy and applicability of flood risk prediction.

[0020] A third aspect of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-mentioned two-dimensional flood risk prediction method.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the two-dimensional flood risk prediction method as described in the above embodiment.

[0022] A fifth aspect of the present invention provides a computer program product, comprising a computer program, which is executed to implement the two-dimensional flood risk prediction method as described in any one of the above items.

[0023] Therefore, the present invention has the following advantages:

[0024] (1) The present invention fully considers the inconsistency characteristics of the hydrological series under the influence of climate change and human underlying surface activities, constructs a hydrological-machine learning coupling model that integrates the attention mechanism to achieve high-reliability simulation of runoff, and constructs a time-varying Copula model considering the inconsistency of the hydrological series. It has strong physical significance and statistical basis and can effectively characterize the changing characteristics of future floods under climate warming.

[0025] (2) The present invention combines a climate multi-model set, a hydrological-machine learning coupling model integrating an attention mechanism, and a most likely combination scenario method with different global warming levels. This can provide an important and highly operational reference for basin flood risk assessment and early warning under a changing environment, and provide engineering reference value for responding to future climate disasters and scientifically formulating emission reduction strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0027] Figure 1 A flow chart of a two-dimensional flood risk prediction method according to an embodiment of the present invention;

[0028] Figure 2 A schematic diagram showing changes in the correlation coefficient between daily measured runoff and simulated runoff at different lag times according to a specific embodiment of the present invention;

[0029] Figure 3 A schematic diagram of the structure of a hydrological-machine learning coupling model provided according to a specific embodiment of the present invention;

[0030] Figure 4 A schematic diagram of the structure of a memory unit of a long short-term memory neural network (LSTM) model provided according to a specific embodiment of the present invention;

[0031] Figure 5 A schematic diagram of the structure of a convolutional attention module (CBAM) according to a specific embodiment of the present invention;

[0032] Figure 6 A flow chart of a two-dimensional flood risk prediction method according to a specific embodiment of the present invention;

[0033] Figure 7 A block diagram of a two-dimensional flood risk prediction device according to an embodiment of the present invention;

[0034] Figure 8 A schematic structural diagram of an electronic device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0036] The following describes the two-dimensional flood risk prediction method, device, electronic device, medium and product of the embodiment of the present invention with reference to the accompanying drawings. In response to the problems of low accuracy and poor applicability of flood risk prediction in related technologies mentioned in the above background technology, the present invention provides a two-dimensional flood risk prediction method, which first collects basic meteorological and hydrological data of the basin, then derives humidity-related data, calibrates the hydrological model and constructs a machine learning model; then, through decoding-encoding bias correction, obtains a meteorological simulation series under the climate change scenario, drives the model to simulate future hydrological processes; extracts flood characteristic values ​​based on the optimal model, constructs a joint probability distribution function; and finally obtains a two-dimensional flood risk prediction result, which solves the problems of low accuracy and poor applicability of flood risk prediction in related technologies and improves the accuracy and applicability of flood risk prediction.

[0037] Specifically, Figure 1This is a flow chart of a two-dimensional flood risk prediction method according to an embodiment of the present invention.

[0038] like Figure 1 As shown in FIG, the two-dimensional flood risk prediction method includes the following steps:

[0039] In step S101 , multimodal data within the target watershed is acquired.

[0040] Among them, the multimodal data in the target basin is a data set with various data types and forms of expression collected through a variety of different sensors, observation methods and data sources within a specifically defined basin area.

[0041] Specifically, the multimodal data includes multiple global climate model simulation data in the target basin, meteorological data from the ERA5 dataset, socioeconomic data, and daily flow series data from the target basin control hydrological station. Among them, the embodiment of the present invention collects the daily flow series from the basin control hydrological station and obtains meteorological data such as precipitation, 2m air temperature, 2m dew point temperature, wind speed, air pressure, shortwave radiation, and longwave radiation from the ERA5 reanalysis dataset.

[0042] For example, this embodiment of the present invention uses a watershed as a research unit, first collecting daily flow series from the watershed's control hydrological stations, and then collecting hourly data from ERA5. ERA5 is the fifth-generation atmospheric reanalysis dataset of the European Centre for Medium-Range Weather Forecasts, with a spatial resolution of 0.25°, providing hourly meteorological data covering the entire globe since 1979. This embodiment obtains hourly precipitation, 2-meter air temperature, 2-meter dew point temperature, wind speed, air pressure, shortwave radiation, and longwave radiation data from the ERA5 dataset for the study area from 1985 to 2014. After time scale conversion, a daily series is obtained, and finally, the basin-averaged daily-scale meteorological series is obtained using the Thiessen polygon method.

[0043] Furthermore, daily meteorological data output by M global climate models (GCMs) were collected. To estimate future climate scenarios, the M global climate models (GCMs) recently released by the Coupled Model Inter-comparison Project Phase 6 (CMIP6) were used. CMIP6 uses a matrix framework of shared socioeconomic pathways (SSPs) and representative concentration pathways (RCPs). The scenarios selected in the present invention include three scenarios for historical and future periods (SSP245, SSP370, and SSP585). The meteorological variables selected are daily precipitation, mean daily temperature, maximum daily temperature, minimum daily temperature, specific humidity, relative humidity, wind speed, shortwave radiation, and longwave radiation data. At the same time, annual potential evapotranspiration data output by the GCMs under the three SSP scenarios were obtained. The historical period was set to 1985-2014, and the future period was set to 2015-2100.

[0044] Furthermore, population, GDP, and land use data, as well as vulnerability indices, were collected from the Shared Socioeconomic Pathway (SSP) dataset for the study basin. To assess the socioeconomic risks posed by flood events, population, GDP, and land use data from three SSPs—moderate development (SSP2), competitive development (SSP3), and conventional development (SSP5)—were considered. These data were combined with the corresponding greenhouse gas emission scenarios (RCPs), using output from three matrix frameworks: SSP245, SSP370, and SSP585. Multiple international institutions provide simulated population and GDP data for SSPs. This paper uses an open-source, policy-based forecast dataset. This product incorporates the results of previous population and economic censuses and annual statistical yearbooks. Based on the Cobb-Douglas model and the population-development-environment analysis model, it estimates socioeconomic indices for the relevant regions from 2010 to 2100. The Land-Use Harmonization2 (LUH2) dataset is a global land use dataset with a spatial resolution of 0.25°. It provides land use patterns, potential land use transitions, and key agricultural management information from 850 to 2100 and is widely used in socioeconomic risk assessment of extreme hydrological events.

[0045] Furthermore, we obtain the vulnerability coefficients of population, GDP and agriculture to floods, which are denoted as Vul POP ,Vul GDP and Vul ALD; Based on this coefficient, socioeconomic risks can be further assessed.

[0046] It should be noted that after obtaining the population, GDP and land use data at the grid scale, the embodiment of the present invention uses the Thiessen polygon method to deduce the average population, GDP and farmland area series of the watershed under the future climate change scenario.

[0047] In step S102, based on the multimodal data in the target watershed, a simulated runoff result is obtained based on multiple pre-established hydrological models, and a hydrological-machine learning coupling model is established based on the simulated runoff result.

[0048] Among them, the multiple pre-established hydrological models are hydrological models established based on the multi-dimensional characteristics of different river basins, and the hydrological-machine learning coupling model is a deep learning model established by considering hydrological data and machine learning models, which is used to predict flood risks.

[0049] In some embodiments, the multimodal data include multiple global climate model simulation data, meteorological data of the ERA5 data set, socioeconomic data and daily flow series data of the target basin control hydrological station. According to the multimodal data in the target basin, the simulated runoff results are obtained based on multiple pre-established hydrological models, and a hydrological-machine learning coupling model is established based on the simulated runoff results, including: calculating the relative humidity and specific humidity results based on the meteorological data of the ERA5 data set; and obtaining the simulated runoff results based on the daily flow series data of the target basin control hydrological station and the meteorological data of the ERA5 data set through multiple preset hydrological models; determining the simulated runoff duration of the machine learning model based on the simulated runoff results and the pre-collected daily measured runoff results according to a preset correlation threshold; and correcting the simulated runoff results based on the long short-term memory neural network of the coupled attention mechanism to obtain the hydrological-machine learning coupling model.

[0050] Specifically, the embodiment of the present invention infers relative humidity and specific humidity based on the meteorological data of the ERA5 dataset, and quantitatively describes the saturated water vapor pressure e by the Clausius-Clapeyron thermodynamic equation. sat Nonlinear relationship with temperature T:

[0051]

[0052] Among them, T0 and e s0 are integral constants, which are 273.16K and 611Pa respectively; L v is the latent heat of vaporization, take 2.5×10 6 Jkg -1 ; R v is the water vapor gas constant, which is 461 J kg -1 K -1 .

[0053] Dew point temperature represents the temperature when air is cooled to water vapor saturation under the conditions of constant water vapor content and air pressure. Substituting it into the Clausius-Clapeyron equation can measure the actual water vapor pressure. 2m ) and dew point temperature (T dew ) are substituted into

[0054] Clausius-Clapeyron thermodynamic equation, deriving the relative humidity near the ground: RH = e sat (T dew ) / e sat (T 2m );

[0055] Specific humidity q is the ratio of water vapor mass to the total mass of the air mass and is derived using ERA5 ground pressure p and dew point temperature:

[0056]

[0057] Furthermore, based on the daily runoff data observed at the hydrological stations and the daily precipitation, daily maximum temperature and daily minimum temperature series of the ERA5 dataset, four hydrological models (including XAJ, GR4J, HBV and SIMHYD) were driven to obtain preliminary simulated runoff. Among them, XAJ is the Xin'anjiang model, GR4J is a conceptual hydrological model developed by the French National Institute of Hydrometeorology (SHMI), which is mainly used for runoff simulation in small and medium-sized watersheds, HBV is a distributed hydrological model developed by the Royal Institute of Technology (KTH) in Sweden, which can be used to simulate hydrological processes in watersheds of different scales, and SIMHYD is a conceptual hydrological model developed by the Commonwealth Scientific and Industrial Research Organization (CSIRO) of Australia.

[0058] Furthermore, statistical analysis was performed on the daily runoff process in the preliminary simulated runoff results and the measured daily runoff process to determine the lag time that affects the daily measured runoff, and an appropriate correlation threshold was selected to determine the simulated runoff duration for establishing a machine learning model with the measured runoff.

[0059] Figure 2 Schematic diagram of the change in correlation coefficient between daily measured runoff and simulated runoff under different lag times according to a specific embodiment of the present invention. Figure 2 As shown in the figure, the correlation coefficient between the simulated runoff and the measured runoff generally decreases with the extension of the lag time. An appropriate correlation threshold is selected, for example, 0.5 is determined as the simulated runoff time of the machine learning model.

[0060] Furthermore, the long short-term memory neural network (Attention-LSTM) model coupled with the attention mechanism was used to correct the preliminary simulated runoff results to quantitatively evaluate the coefficient T sThe highest is the objective function, and the fitting model is calibrated to establish a hydrological-machine learning coupling model.

[0061] Figure 3 This is a schematic diagram of the hydrology-machine learning coupling model structure provided according to a specific embodiment of the present invention. Figure 3 As shown, a long short-term memory neural network (LSTM) model with a three-layer neural network architecture is constructed to generalize the regulating and storage effects of dams, reservoirs or water diversion projects on watersheds and improve the accuracy of hydrological simulation. This embodiment uses a neural network interval simulation mean method to independently run the neural network model multiple times and take the average value as the final simulation result to reduce uncertainty.

[0062] Furthermore, if Figure 4 As shown, Figure 4 A schematic diagram of the structure of a long short-term memory neural network (LSTM) model memory unit provided according to a specific embodiment of the present invention. In order to solve the gradient explosion and gradient vanishing problems caused by the nonlinear autoregressive exogenous input model (NARX) dynamic neural network in the deep learning process (the number of hidden layers ≥ 2 layers), the LSTM long short-term memory neural network introduces storage units, namely input gates, forget gates, internal feedback connections, and output gates in the hidden layers of the NARX neural network to select memory current information or forget past memory information (such as rainfall-runoff mapping relationship) to enhance the long-term memory capacity of the NARX neural network. In short, the LSTM long short-term memory neural network replaces each hidden layer in the NARX dynamic neural network with a storage unit with memory function, referred to as an LSTM unit, wherein the input layer and output layer are the same as those of the NARX dynamic neural network.

[0063] Furthermore, in order to better focus on key meteorological variables that affect runoff simulation and improve simulation accuracy, the embodiment of the present invention adds an Attention layer to the LSTM to enhance the context modeling capability of the LSTM when processing time series data. Figure 5 A schematic diagram of the structure of a convolutional attention module (CBAM) according to a specific embodiment of the present invention is shown in FIG. Figure 5 As shown, the convolutional attention module (CBAM) of the embodiment of the present invention can dynamically allocate attention according to different time steps of the input, so as to better capture long-term dependencies and key information. CBAM is a modular attention mechanism, including channel attention and spatial attention. For the output of LSTM, the Attention layer first calculates the channel attention. The role of the channel attention module is to assign a weight to the features of each time step according to the importance of each channel. The average pooling and maximum pooling formulas are:

[0064] M avg =AvgPool(Ht );

[0065] M max =MaxPool(H t );

[0066] Among them, AvgPool and MaxPool are global average pooling and global maximum pooling respectively, H t is the output feature (i.e., hidden state) of LSTM at time t.

[0067] Furthermore, the two pooling results are concatenated and passed to a shared fully connected layer to obtain the channel attention weight:

[0068] M channel =σ(W ch [M avg ,M max ]+b ch ):

[0069] Among them, W ch and b ch represent weights and biases respectively, and σ is the sigmoid activation function.

[0070] Furthermore, for each time step feature, the spatial attention module enhances the important locations that the model focuses on by weighting each spatial location. After performing average pooling and max pooling on the channel dimension, the results are concatenated and a convolution operation is performed to generate a spatial attention map. The output of the convolution layer is passed through the sigmoid activation function to generate the spatial attention weights:

[0071] M spatial =σ(Conv([S avg ,S max ]));

[0072] Among them, S avg To perform average pooling results in the channel dimension, S max To perform the maximum pooling result in the channel dimension, Conv represents the convolution operation and σ is the sigmoid activation function.

[0073] Furthermore, channel attention and spatial attention are applied to the hidden state sequence of LSTM respectively. Meteorological data such as daily precipitation, daily average temperature, daily maximum temperature, daily minimum temperature, specific humidity, relative humidity, wind speed, shortwave radiation, and longwave radiation data obtained from ERA products, as well as simulated runoff series and measured runoff series are used as input. After calibrating the LSTM model, an Attention layer is connected to further calibrate the simulated runoff series. The equation is:

[0074] R cor (t) = FAttention-LSTM [QM(t),QM(t-1),QM(t-2),…,QM(tN)];

[0075] Among them, R cor (t) represents the corrected runoff at time t, QM(t) represents the input variables for calibrating the Attention-LSTM model, including the daily runoff series simulated by the four hydrological models and the basin-averaged meteorological data derived by ERA5; QM(t-1) represents the simulated runoff and meteorological series at time t-1, and N represents the lag determined by the Attention-LSTM model; F Attention-LSTM Represents the Attention-LSTM model.

[0076] Furthermore, the Attention-LSTM model is trained using the minimum batch gradient descent method to optimize the model parameters to quantitatively evaluate the coefficient T s The highest target calibration fitted model:

[0077]

[0078] Where KGE() is the Kling-Gupta efficiency coefficient, Q obs and Q sim represents the daily runoff data of the simulated and measured series; and Represents the monthly flood duration data of the simulated series and the measured series; and represents the flood flow data of the simulated series and the measured series at the monthly scale; α, β, and γ represent their respective weight parameters, satisfying α+β+γ=1. s The value range is (-∞,1], when T s =1, it means that the simulated series is completely consistent with the measured series.

[0079] Therefore, the embodiment of the present invention establishes a hydrological-machine learning coupling model by coupling the hydrological model and the machine learning model, and the model is denoted as HLM.

[0080] In step S103, the basin hydrological process under the future scenario is simulated by the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under the future scenario. Based on the simulation results of the basin hydrological process, the preset optimal model is used to extract the flood duration and flood volume characteristic values ​​under multiple warming level time windows. Based on the flood duration and flood volume characteristic values ​​under multiple warming level time windows, a two-dimensional flood risk prediction result is obtained.

[0081] Among them, the hydrological-machine learning coupling model is a machine learning model of the hydrological process. The future scenario is based on a preset time point. Based on current research and analysis, it is a hypothesis or simulation of the climate change, socio-economic development and other conditions that the basin may face in the future after the node. Flood duration and flood volume are two important characteristic quantities in the flood process, and their marginal distribution functions are used to describe the probability distribution of their respective variables.

[0082] Furthermore, in some embodiments, the basin hydrological process under future scenarios is simulated by a hydrological-machine learning coupling model to obtain simulation results of the basin hydrological process under future scenarios, including: based on the corrected meteorological simulation series results, the basin hydrological process under future scenarios is simulated by a hydrological-machine learning coupling model to obtain a series of daily runoff results under future climate change scenarios; based on the daily runoff series results under future climate change scenarios, the basin hydrological process simulation results under future scenarios are obtained.

[0083] Specifically, the corrected meteorological simulation series results are input into the hydrological-machine learning coupling model to simulate the daily runoff series under future climate change scenarios and obtain the simulation results of the basin hydrological processes under future scenarios.

[0084] Furthermore, in some embodiments, before extracting flood duration and flood volume characteristic values ​​under multiple warming level time windows using a preset optimal model based on the simulation results of the basin hydrological process, it also includes: calculating the time windows under different warming levels based on the temperature data in the corrected meteorological simulation series results; fusing the corrected meteorological simulation series results with the machine learning model to obtain a fused machine learning model; determining the model parameters of the fused machine learning model based on a preset adaptive model selection mechanism, and obtaining the preset optimal model according to the determination results.

[0085] Specifically, taking 1985-2014 as the historical period and 2015-2100 as the future period, the temperature rise relative to the pre-industrial level in a future year is defined as:

[0086]

[0087] in, T is the global average temperature in the future year t averaged by the ensemble of M global climate models, his is the global average temperature during the historical period, ΔT his is the temperature rise over the historical period relative to pre-industrial levels.

[0088] Furthermore, ΔT(t 1.5℃ )=1.5℃、ΔT(t 2.0℃ )=2.0℃ and ΔT(t3.0℃ )=year t when 3.0℃ 1.5℃ , t 2.0℃ and t 3.0℃ , and use this year as the midpoint year to obtain a 30-year time window under different warming levels.

[0089] Furthermore, a series of meteorological simulations under M climate change scenarios were combined with a machine learning model (GCM-HLM). The optimal model was dynamically selected based on an adaptive model selection mechanism. Considering the combination of M meteorological simulations and four HLMs, a baseflow separation method was used to extract the duration and volume of measured runoff and simulated runoff for each simulated basin hydrological process. The error measurement formula between the two was defined as:

[0090]

[0091] Among them, DTW stands for Dynamic Time Warping algorithm, which is used to calculate the similarity between time series. obs (t) and represent the duration of the measured runoff and the simulated runoff output by the i-th hydrological machine learning model under the k-th global climate model, respectively, and S obs (t) and represent the measured runoff and the simulated runoff output by the i-th hydrological machine learning model under the k-th global climate model, respectively, and n is the time step.

[0092] Furthermore, considering the duration error and flood error comprehensively, the total error evaluation function is obtained as follows:

[0093]

[0094] Among them, α is a weight parameter, which is used to adjust the relative importance of duration error and flood error.

[0095] Furthermore, the optimal model is adaptively selected based on the minimum error:

[0096]

[0097] Furthermore, in some embodiments, based on the characteristic values ​​of flood duration and flood volume under multiple warming level time windows, a two-dimensional flood risk prediction result is obtained, including: constructing a marginal distribution function of flood duration and flood volume based on the characteristic values ​​of flood duration and flood volume under multiple warming level time windows, and constructing a Copula joint distribution function of flood duration and flood volume under inconsistency conditions; based on the marginal distribution function of flood duration and flood volume and the Copula joint distribution function of flood duration and flood volume under inconsistency conditions, calculating the joint recurrence period of flood duration and flood volume under different warming level time windows to obtain a two-dimensional flood risk prediction result.

[0098] Among them, the Copula joint distribution function of flood duration and flood volume under inconsistent conditions is the combined distribution function that combines the inconsistent marginal distribution function of flood duration and the marginal distribution function of flood volume. The joint return period of flood duration and flood volume is the combination with the largest joint probability density function on the return period contour line. The two-dimensional flood risk prediction results include the socioeconomic risk prediction results of increased future flood risks under various warming levels.

[0099] Specifically, the 95% daily runoff quantile is used to identify the annual maximum flood process; the duration and flood volume of the flood process are calculated, and X is set to represent the flood characteristic quantities (duration D and flood volume S). The present invention uses the gamma distribution, generalized extreme value distribution, inverse Gaussian distribution, and Weibull distribution functions as candidate marginal distribution functions for flood duration and flood volume, and uses the Kolmogorov-Smirnov (KS) statistic for goodness of fit test:

[0100]

[0101] Among them, F n (x) represents the observed sequence value, F n (x) represents the theoretical sequence value.

[0102] Furthermore, the marginal distribution function is optimized based on the AIC criterion:

[0103] AIC = 2k-2ln(L);

[0104] Where k is the number of model parameters and L is the likelihood function.

[0105] Furthermore, the embodiment of the present invention uses Gumbel copula, Frank copula, Clayton copula, t-copula, and Gaussian copula as candidate joint distribution functions for flood duration and flood volume. After using the KS goodness-of-fit test, the copula joint distribution function of flood duration and flood volume under inconsistent conditions is constructed according to the AIC criterion. Based on the definition of the copula function, the inconsistent two-variable copula function can be expressed as:

[0106]

[0107] Among them, F t (d t ,s t ) represents the time-varying joint distribution function of D and S; and denote the time-varying marginal distribution function and time-varying parameters of the D and S variables, respectively.

[0108] Furthermore, the OR return period is used as a flood metric, which is defined as:

[0109]

[0110] in, is the joint return period of time-varying OR, in years.

[0111] Furthermore, the most likely combination of flood duration and flood volume is the combination with the largest joint probability density function on the return period contour line (d * (t),s * (t)), and solve it by constructing the following equation:

[0112]

[0113] Among them, f t (d t ,s t ) represents the density function of the time-varying joint distribution function of duration D and flood volume S; The density function representing the time-varying Copula joint distribution function; and Respectively and The density function of the marginal distribution function.

[0114] Thus, given the joint return period T or , the above method is used to obtain the historical period (1985-2014) in T or After the most likely combination of flood duration and flood volume in each year under the given conditions, the characteristic values ​​of flood duration and flood volume in the historical period were obtained.

[0115] Furthermore, the obtained flood duration and flood volume characteristic values ​​in the historical period were substituted into the most likely combination model of different future warming periods to calculate the new return period T for the 30-year window period under different warming levels. f (ΔT), the given historical period return period is recorded as T h , the future socioeconomic risk corresponding to the global temperature rise ΔT can be measured by the following formula:

[0116]

[0117] Among them, E pop 、E GDP and E ALD Respectively represent the population, GDP and farmland risks affected by floods, POP k , GDP k and ALD kare the population, GDP and farmland area in the kth year within the 30-year window corresponding to the global temperature rise ΔT; I(·) is the indicator function, T h -T f When (ΔT)>0, it is recorded as 1, indicating that the two-dimensional flood risk increases, otherwise it is recorded as 0, indicating that the risk decreases; N1 and N2 represent the start and end years of the study period, respectively.

[0118] Furthermore, the comprehensive socioeconomic risks under different warming levels are evaluated and predicted. Based on the socioeconomic risk analysis results of different warming levels, the comprehensive socioeconomic risks under the warming level corresponding to the global temperature rise ΔT are:

[0119]

[0120] Among them, SocioRisk represents the integrated socioeconomic risk; E POP (max), E GDP (max) and E ALD (max) represents the maximum values ​​of population, GDP and farmland risks affected by floods obtained under M global climate models.

[0121] Furthermore, in some embodiments, before simulating the basin hydrological process under future scenarios through the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under future scenarios, it also includes: based on the decoding-encoding deviation correction method, correcting multiple global climate model simulation data to obtain a corrected meteorological simulation series results.

[0122] Specifically, the observation data of each month in the ERA5 reanalysis dataset from 1985 to 2014 are recorded as The simulated data of GCMs for each month in the historical period are recorded as The simulated data of GCMs for each month in the future period of a certain scenario is recorded as Separate deduction and The interannual trend line of the series is obtained by decoding the trend term caused by anthropogenic climate forcing; the trend line is then shifted so that the sum of the trend terms of each year is 0; for the series with insignificant interannual trends at the 0.05 confidence level, the trend terms of each year are reset to 0; finally, the trend terms of each series are removed to obtain three new series. and This is the data series caused by internal climate variability.

[0123] Furthermore, we calculate the historical period under different quantiles and scenario period The deviation of the pseudo observation series of the scenario period is obtained based on the real observation values.

[0124]

[0125] Where p is the set quantile, and 100 equally spaced quantiles are taken in the range of 0.01 to 0.99.

[0126] Further, based on Scenario simulation series The frequency distribution curve is used to correct the deviation, and the corrected future series is obtained based on quantile mapping; finally, the coding method is used to superimpose the trend term deducted in the first step to obtain the corrected future scenario.

[0127] Therefore, the trend of each month is retained, and the bias correction method is applied on the daily scale. Considering that the frequency distribution functions of each month may vary to a certain extent, the data of each GCM are corrected for 12 months. The daily series of meteorological data of M GCMs under three scenarios (SSP245, SSP370 and SSP585) in the historical period and the future period are obtained, and finally the corrected meteorological simulation series results are obtained.

[0128] In order to enable relevant practitioners in this field to better understand the two-dimensional flood risk prediction method of the embodiment of the present invention, the two-dimensional flood risk prediction method will be explained below with reference to specific embodiments.

[0129] Figure 6 This is a flow chart of a two-dimensional flood risk prediction method according to a specific embodiment of the present invention.

[0130] like Figure 6 As shown, the embodiment of the present invention first collects basic meteorological and hydrological data of the basin, then derives relative humidity and specific humidity, calibrates the basin hydrological model and machine learning model, and then obtains M sets of meteorological simulation series under climate change scenarios based on the decoding-encoding bias correction method, and drives the hydrological-machine learning coupling model integrating the attention mechanism to simulate the basin hydrological process under future scenarios; based on the corrected global climate model set, the time windows under different warming levels are deduced, and the optimal model is dynamically selected according to the adaptive model selection mechanism; based on the simulated hydrological process and the annual maximum sampling method, the characteristic values ​​of flood duration and flood volume under different warming level time windows are extracted, and a joint probability distribution function based on Copula under inconsistency conditions is established; based on the most likely combination scenario, the joint recurrence period of flood duration and flood volume under inconsistency conditions is obtained, and the impact of climate change and underlying human activities on the future flood situation in the basin is evaluated; finally, based on the shared socioeconomic path dataset, the integrated socioeconomic risk of increased flood risk in the future is deduced.

[0131] According to the two-dimensional flood risk prediction method provided by the embodiment of the present invention, basic meteorological and hydrological data of the watershed are first collected, and then humidity-related data are derived, the hydrological model is calibrated and a machine learning model is constructed; then, a meteorological simulation series under the climate change scenario is obtained through decoding-encoding bias correction, and the model is driven to simulate future hydrological processes; flood characteristic values ​​are extracted based on the optimal model, and a joint probability distribution function is constructed; finally, a two-dimensional flood risk prediction result is obtained, which solves the problems of low flood risk prediction accuracy and poor applicability in related technologies and improves the accuracy and applicability of flood risk prediction.

[0132] Next, a two-dimensional flood risk prediction device according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0133] Figure 7 Schematic diagram of a two-dimensional flood risk prediction device according to an embodiment of the present invention.

[0134] like Figure 7 As shown, the two-dimensional flood risk prediction device 10 includes: an acquisition module 100 , a mathematical modeling module 200 and a prediction module 300 .

[0135] Among them, the acquisition module 100 is used to obtain multimodal data in the target basin; the mathematical modeling module 200 is used to obtain simulated runoff results based on multiple pre-established hydrological models according to the multimodal data in the target basin, and establish a hydrological-machine learning coupling model based on the simulated runoff results; the prediction module 300 is used to simulate the basin hydrological process under future scenarios through the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under future scenarios, and based on the basin hydrological process simulation results, use the preset optimal model to extract the flood duration and flood volume characteristic values ​​under multiple warming level time windows, and obtain a two-dimensional flood risk prediction result based on the flood duration and flood volume characteristic values ​​under multiple warming level time windows.

[0136] Furthermore, in some embodiments, the mathematical modeling module 200 is specifically used to: calculate relative humidity and specific humidity results based on the meteorological data of the ERA5 data set; and obtain simulated runoff results through multiple preset hydrological models based on the daily flow series data of the target watershed control hydrological station and the meteorological data of the ERA5 data set; determine the simulated runoff duration of the machine learning model based on the simulated runoff results and the pre-collected daily measured runoff results according to a preset correlation threshold; correct the simulated runoff results based on the long short-term memory neural network with a coupled attention mechanism to obtain a hydrological-machine learning coupling model.

[0137] Furthermore, in some embodiments, before simulating the basin hydrological process under future scenarios through the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under future scenarios, the prediction module 300 is also used to: based on the decoding-encoding bias correction method, correct multiple global climate model simulation data to obtain a corrected meteorological simulation series results.

[0138] Furthermore, in some embodiments, the prediction module 300 is also used to: simulate the basin hydrological process under future scenarios through a hydrological-machine learning coupling model based on the corrected meteorological simulation series results, and obtain a series of daily runoff results under future climate change scenarios; based on the daily runoff series results under future climate change scenarios, obtain basin hydrological process simulation results under future scenarios.

[0139] Furthermore, in some embodiments, before extracting flood duration and flood volume characteristic values ​​under multiple warming level time windows using a preset optimal model based on the simulation results of the basin hydrological process, the prediction module 300 is also used to: calculate the time windows under different warming levels based on the temperature data in the corrected meteorological simulation series results; fuse the corrected meteorological simulation series results with the machine learning model to obtain a fused machine learning model; determine the model parameters of the fused machine learning model based on a preset adaptive model selection mechanism, and obtain the preset optimal model based on the determination results.

[0140] Furthermore, in some embodiments, the prediction module 300 is also used to: construct a marginal distribution function of flood duration and flood volume based on the characteristic values ​​of flood duration and flood volume under multiple warming level time windows, and construct a Copula joint distribution function of flood duration and flood volume under inconsistent conditions; calculate the joint recurrence period of flood duration and flood volume under different warming level time windows based on the marginal distribution function of flood duration and flood volume and the Copula joint distribution function of flood duration and flood volume under inconsistent conditions, and obtain a two-dimensional flood risk prediction result.

[0141] It should be noted that the above explanations of the embodiment of the two-dimensional flood risk prediction method are also applicable to the two-dimensional flood risk prediction device of this embodiment, and will not be elaborated here.

[0142] According to the two-dimensional flood risk prediction device provided by the embodiment of the present invention, basic meteorological and hydrological data of the watershed are first collected, and then humidity-related data are derived, the hydrological model is calibrated and a machine learning model is constructed; then, a meteorological simulation series under a climate change scenario is obtained through decoding-encoding bias correction, and the model is driven to simulate future hydrological processes; flood characteristic values ​​are extracted based on the optimal model, and a joint probability distribution function is constructed; finally, a two-dimensional flood risk prediction result is obtained, which solves the problems of low flood risk prediction accuracy and poor applicability in related technologies and improves the accuracy and applicability of flood risk prediction.

[0143] Figure 8 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present invention. The electronic device may include:

[0144] A memory 801 , a processor 802 , and a computer program stored in the memory 801 and executable on the processor 802 .

[0145] When the processor 802 executes the program, the two-dimensional flood risk prediction method provided in the above embodiment is implemented.

[0146] Furthermore, the electronic device further includes:

[0147] The communication interface 803 is used for communication between the memory 801 and the processor 802 .

[0148] The memory 801 is used to store computer programs that can be run on the processor 802.

[0149] The memory 801 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0150] If the memory 801, processor 802, and communication interface 803 are implemented independently, the communication interface 803, memory 801, and processor 802 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0151] Optionally, in a specific implementation, if the memory 801, the processor 802 and the communication interface 803 are integrated on a chip, the memory 801, the processor 802 and the communication interface 803 can communicate with each other through an internal interface.

[0152] The processor 802 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0153] In addition, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the two-dimensional flood risk prediction method according to the above embodiment is implemented.

[0154] In addition, an embodiment of the present invention further provides a computer program product, including a computer program, which is executed to implement the two-dimensional flood risk prediction method as described in the above embodiment.

[0155] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples without contradiction.

[0156] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "N" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0157] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing a custom logical function or step of a process, and the scope of the preferred embodiments of the invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0158] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiment, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0159] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

Claims

1. A two-dimensional flood risk prediction method, characterized in that: The following steps are involved: Acquire multimodal data within the target watershed; According to the multimodal data in the target watershed, simulated runoff results are obtained based on multiple pre-established hydrological models, and a hydrological-machine learning coupling model is established based on the simulated runoff results; The hydrological process of the river basin under the future scenario is simulated by the hydrological-machine learning coupling model to obtain the simulation results of the river basin hydrological process under the future scenario. Based on the simulation results of the river basin hydrological process, the preset optimal model is used to extract the flood duration and flood volume characteristic values ​​under multiple warming level time windows. Based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows, a two-dimensional flood risk prediction result is obtained.

2. The two-dimensional flood risk prediction method according to claim 1, characterized in that: The multimodal data includes multiple global climate model simulation data, meteorological data of the ERA5 dataset, socioeconomic data, and daily flow series data of the target basin control hydrological station. The simulated runoff results are obtained based on the multimodal data in the target basin and multiple pre-established hydrological models, and a hydrological-machine learning coupling model is established based on the simulated runoff results, including: Relative humidity and specific humidity results are calculated based on the meteorological data of the ERA5 dataset; and the simulated runoff results are obtained through multiple preset hydrological models based on the daily flow series data of the target watershed control hydrological station and the meteorological data of the ERA5 dataset; Based on the simulated runoff results and the pre-collected daily measured runoff results, and according to a preset correlation threshold, determining the simulated runoff duration of the machine learning model; Based on a long short-term memory neural network with a coupled attention mechanism, the simulated runoff results are corrected to obtain the hydrological-machine learning coupling model.

3. The two-dimensional flood risk prediction method according to claim 1, characterized in that: Before simulating the basin hydrological process under the future scenario by the hydrological-machine learning coupling model to obtain the basin hydrological process simulation result under the future scenario, the method further includes: Based on the decoding-encoding deviation correction method, the plurality of global climate model simulation data are corrected to obtain a corrected series of meteorological simulation results.

4. The two-dimensional flood risk prediction method according to claim 3, characterized in that: The hydrological process of the watershed under the future scenario is simulated by the hydrological-machine learning coupling model to obtain the simulation results of the watershed hydrological process under the future scenario, including: Based on the corrected meteorological simulation series results, the hydrological process of the watershed under the future scenario is simulated by the hydrological-machine learning coupling model to obtain the daily runoff series results under the future climate change scenario; Based on the daily runoff series results under the future climate change scenario, the simulation results of the basin hydrological process under the future scenario are obtained.

5. The two-dimensional flood risk prediction method according to claim 3, characterized in that: Before extracting flood duration and flood volume characteristic values ​​under multiple warming level time windows based on the basin hydrological process simulation results using a preset optimal model, the method further includes: Calculating time windows at different warming levels based on temperature data in the corrected meteorological simulation series results; fusing the corrected meteorological simulation series results with the machine learning model to obtain a fused machine learning model; Based on a preset adaptive model selection mechanism, the model parameters of the fused machine learning model are determined, and the preset optimal model is obtained according to the determination result.

6. The two-dimensional flood risk prediction method according to claim 1, characterized in that: The two-dimensional flood risk prediction results are obtained based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows, including: Based on the characteristic values ​​of flood duration and flood volume under the multiple warming level time windows, constructing marginal distribution functions of flood duration and flood volume, and constructing a copula joint distribution function of flood duration and flood volume under inconsistency conditions; Based on the marginal distribution function of the flood duration and flood volume and the Copula joint distribution function of the flood duration and flood volume under the inconsistency condition, the joint recurrence period of the flood duration and flood volume under different warming level time windows is calculated to obtain the two-dimensional flood risk prediction result.

7. A two-dimensional flood risk prediction device, characterized in that: The device comprises: Acquisition module, used to obtain multimodal data in the target watershed; a mathematical modeling module, configured to obtain simulated runoff results based on the multimodal data in the target watershed and a plurality of pre-established hydrological models, and to establish a hydrological-machine learning coupling model based on the simulated runoff results; The prediction module is used to simulate the basin hydrological process under the future scenario through the hydrological-machine learning coupling model to obtain the simulation results of the basin hydrological process under the future scenario, and based on the simulation results of the basin hydrological process, use a preset optimal model to extract the flood duration and flood volume characteristic values ​​under multiple warming level time windows, and obtain a two-dimensional flood risk prediction result based on the flood duration and flood volume characteristic values ​​under the multiple warming level time windows.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the two-dimensional flood risk prediction method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the two-dimensional flood risk prediction method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the two-dimensional flood risk prediction method according to any one of claims 1 to 6 is implemented.

Images