Multi-dimensional hydrological risk assessment method and device and computer equipment

Abstract

The invention relates to a multi-dimensional hydrological risk assessment method and device and computer equipment. The method comprises the following steps: acquiring hydrological real-time data and hydrological historical data of a target water area, and respectively inputting the hydrological real-time data and the hydrological historical data into a hydrological risk machine model and a hydrological risk physical model to obtain prediction data of the hydrological real-time data and the hydrological historical data; and if the difference between the two prediction data exceeds a preset value, adjusting the model parameters, inputting the data again, and performing prediction until the prediction difference is smaller than a set threshold value. And then selecting a model result with relatively small difference as water area risk prediction data, performing availability check on the prediction data, and if the data is credible, using the data as target risk prediction data for subsequent application. By adopting the method, the accuracy and timeliness of a hydrological risk prediction result can be effectively improved, and more accurate and real-time data support is provided for water area management, disaster early warning and decision making.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a multi-dimensional hydrological risk assessment method, apparatus and computer equipment. Background Art

[0002] In traditional technologies, hydrological risk assessment technology usually uses remote sensing technology to obtain surface change data, combined with weather forecasts and historical precipitation data, and uses hydrological models to simulate precipitation, evaporation, soil infiltration and other processes to predict water flow dynamics in the basin. And further, by comparing hydrological data under different scenarios, possible hydrological disasters such as flood risks and drought risks are assessed, and hydrological risk assessment data is given. However, traditional technologies often rely on limited observation data and model assumptions, and are difficult to cope with complex hydrological changes and extreme weather events, resulting in poor accuracy and timeliness of hydrological risk prediction results. Summary of the invention

[0003] Based on this, it is necessary to provide a multi-dimensional hydrological risk assessment method, device and computer equipment that can effectively improve the accuracy and timeliness of hydrological risk prediction results in response to the above-mentioned technical problems.

[0004] First, this application provides a multi-dimensional hydrological risk assessment method, including:

[0005] Obtain real-time and historical hydrological data of the target waters;

[0006] Inputting the real-time hydrological data and the historical hydrological data into the hydrological risk machine model and the hydrological risk physical model respectively, to obtain risk machine model prediction data and risk physical model prediction data;

[0007] When the difference between the risk machine model prediction data and the risk physical model prediction data is greater than a preset difference, adjusting the model parameters of the hydrological risk machine model and the hydrological risk physical model according to the difference;

[0008] The adjusted hydrological risk machine model is used as the hydrological risk machine model, and the adjusted hydrological risk physical model is used as the hydrological risk physical model, and the step of inputting the hydrological real-time data and the hydrological historical data into the hydrological risk machine model and the hydrological risk physical model respectively to obtain the risk machine model prediction data and the risk physical model prediction data is returned to execute;

[0009] Until the difference between the risk machine model prediction data and the risk physical model prediction data is less than the preset difference, the risk machine model prediction data or the risk physical model prediction data is selected as the water area risk prediction data;

[0010] An availability check is performed on the water area risk prediction data, and when a check result of the availability check indicates that the water area risk prediction data is credible, the water area risk prediction data is used as target risk prediction data.

[0011] In the second aspect, the present application also provides a multi-dimensional hydrological risk assessment device, including:

[0012] A hydrological data acquisition module is used to obtain real-time and historical hydrological data of the target water area;

[0013] A hydrological data analysis module, used to input the real-time hydrological data and the historical hydrological data into a hydrological risk machine model and a hydrological risk physical model, respectively, to obtain risk machine model prediction data and risk physical model prediction data;

[0014] A hydrological model adjustment module, configured to adjust the model parameters of the hydrological risk machine model and the hydrological risk physical model according to the difference between the risk machine model prediction data and the risk physical model prediction data when the difference is greater than a preset difference;

[0015] The hydrological data analysis module is further used to use the adjusted hydrological risk machine model as the hydrological risk machine model, and the adjusted hydrological risk physical model as the hydrological risk physical model, and return to execute the step of inputting the hydrological real-time data and the hydrological historical data into the hydrological risk machine model and the hydrological risk physical model respectively to obtain the risk machine model prediction data and the risk physical model prediction data;

[0016] A prediction data selection module, used for selecting the risk machine model prediction data or the risk physical model prediction data as the water area risk prediction data when the difference between the risk machine model prediction data and the risk physical model prediction data is less than the preset difference;

[0017] The risk data determination module is used to perform an availability check on the water area risk prediction data, and when the check result of the availability check indicates that the water area risk prediction data is credible, use the water area risk prediction data as target risk prediction data.

[0018] In a third aspect, the present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements any step of a multi-dimensional hydrological risk assessment when executing the computer program.

[0019] The above-mentioned multi-dimensional hydrological risk assessment method, device and computer equipment input the real-time hydrological data and historical hydrological data of the target water area into the hydrological risk machine model and the hydrological risk physical model, and compare the prediction results of the two models. When the difference exceeds the preset threshold, the physical parameters in the model will be automatically adjusted to reduce the model error; after multiple adjustments and iterations, the most credible risk prediction data can be finally selected to ensure the effectiveness of the prediction results in practical applications. Further availability checks of water area risk prediction data further enhance the credibility of the model output and avoid decision-making errors caused by model errors or data inconsistencies. Due to the integration of the advantages of machine learning and physical modeling, the scientificity and practicality of water area risk prediction are improved, so it can effectively improve the accuracy and timeliness of hydrological risk prediction results, and provide more accurate and real-time data support for water area management, disaster warning and decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 An application environment diagram of a multi-dimensional hydrological risk assessment method in an embodiment;

[0022] Figure 2 A schematic diagram of a flow chart of a multi-dimensional hydrological risk assessment method in one embodiment;

[0023] Figure 3 A schematic diagram of a process for obtaining risk physics model prediction data in one embodiment;

[0024] Figure 4 A schematic diagram of a process for obtaining target risk prediction data in one embodiment;

[0025] Figure 5 A schematic diagram of a process for obtaining risk data credibility analysis results in one embodiment;

[0026] Figure 6 A schematic flow chart of a method for obtaining risk data credibility analysis results in another embodiment;

[0027] Figure 7 is a structural block diagram of a multi-dimensional hydrological risk assessment device in one embodiment;

[0028] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0030] The multi-dimensional hydrological risk assessment method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0031] In an exemplary embodiment, Figure 2 As shown in the figure, a multi-dimensional hydrological risk assessment method is provided. Figure 1 The server in the example is used to illustrate, including the following steps 202 to 212. Among them:

[0032] Step 202, obtaining real-time hydrological data and historical hydrological data of the target water area.

[0033] The target water area can be a specific water body or water area for hydrological risk prediction. This can be a river, lake, river basin or reservoir, etc. It is usually necessary to monitor the hydrological conditions of the area in order to predict and manage possible floods, droughts or other hydrological risks.

[0034] Among them, real-time hydrological data can be data on various hydrological parameters (such as water level, flow rate, precipitation, temperature, etc.) of the target water area collected in real time at a specific time point or time period.

[0035] The hydrological historical data may be the hydrological data related to the target water area recorded and saved over a period of time in the past, such as historical water level, flow, precipitation, etc.

[0036] Specifically, the real-time hydrological data of the target water area monitors variables such as water level, flow rate, precipitation, temperature, and wind speed in real time through sensors, remote sensing equipment, meteorological stations, and other equipment installed around the water area. The historical hydrological data of the target water area comes from institutions such as the Meteorological Bureau and hydrological monitoring stations, and usually contains records of hydrological events over many years, such as precipitation history, flood events in the basin, drought conditions, historical water levels, historical flows, and historical precipitation.

[0037] Step 204 , inputting the real-time hydrological data and the historical hydrological data into the hydrological risk machine model and the hydrological risk physical model respectively, to obtain the risk machine model prediction data and the risk physical model prediction data.

[0038] Among them, the hydrological risk machine model can be a model based on machine learning or data mining technology, which is used to analyze and predict the hydrological risks of water areas, and is obtained by training the model to identify laws and patterns in historical data.

[0039] Among them, the hydrological risk physical model can be a model that simulates the hydrological process of a water area based on the principles of hydrology and meteorology using physical equations and laws (such as water flow equations, precipitation-runoff models, etc.). It simulates physical processes such as water flow, precipitation, evaporation, etc. in a water area through mathematical modeling and numerical calculations to assess hydrological risks. Unlike machine models, physical models focus on the physical mechanisms and laws of the real world.

[0040] Among them, the risk machine model prediction data can be the prediction results obtained by the hydrological risk machine learning model. These data usually include the estimation of hydrological variables (such as water level, flow, etc.) in the future, and the risk assessment results (such as flood risk, drought risk, etc.) calculated based on these prediction data.

[0041] Among them, the risk physical model prediction data can be the prediction results obtained through the hydrological risk physical model. These data simulate the hydrological process and physical laws, calculate the hydrological state (such as water level, flow rate, etc.) of the water area and its changing trend, and conduct risk assessment based on these prediction results.

[0042] Specifically, the real-time hydrological data and the historical hydrological data need to be input into two different types of models respectively. The hydrological risk machine model (such as support vector machine, decision tree, random forest, neural network, etc.) predicts hydrological risks based on the real-time hydrological data and the historical hydrological data by learning the patterns in the historical data, especially those complex nonlinear relationships. The hydrological risk physical model is based on the basic principles of hydrology and meteorology. According to the real-time hydrological data and the historical hydrological data, the hydrodynamic equations (such as precipitation-runoff model, surface flow model, etc.) are used to simulate the physical change process of the water body. These two types of models predict hydrological risks from different perspectives, and obtain the risk machine model prediction data and the risk physical model prediction data.

[0043] Step 206 , when the difference between the risk machine model prediction data and the risk physical model prediction data is greater than a preset difference, the model parameters of the hydrological risk machine model and the hydrological risk physical model are adjusted according to the difference.

[0044] Among them, the difference can be the deviation or difference between the hydrological risk machine model prediction data and the hydrological risk physical model prediction data, and the difference is used to measure the similarity or consistency of the outputs of the two models.

[0045] The preset difference can be a tolerance error threshold set in advance during the model adjustment process. It is used to measure the size of the difference between the prediction results of the machine learning model and the physical model. If the difference between the prediction results of the two models exceeds this threshold, it means that the model predictions are inconsistent and the model parameters need to be further adjusted; if the difference is within the threshold range, it is considered that the predictions of the two models are sufficiently consistent.

[0046] Specifically, if the difference between the prediction results of the risk machine model prediction data and the risk physical model prediction data is greater than the preset tolerance threshold, it means that the prediction accuracy of one or both models needs to be improved. In this case, it is necessary to adjust the parameters of the model according to the difference in the prediction results to narrow the prediction difference between the two models. For the hydrological risk machine model, the model's fitting ability is optimized and overfitting or underfitting is reduced by adjusting the model's hyperparameters (such as learning rate, number of hidden layer nodes, regularization terms, etc.); for the hydrological risk physical model, the focus of adjustment is on the parameters involved in the physical equations, such as the friction coefficient of water flow, precipitation-runoff conversion coefficient, etc., to more accurately simulate the actual hydrological process. Since there may be conflicting adjustments during the adjustment process of the two models, optimization algorithms such as particle swarm optimization (PSO), genetic algorithm (GA) or Bayesian optimization are used to automatically explore the parameter space and find parameter combinations that can optimize the prediction results of the two models at the same time. The optimization process is usually iterative. By repeatedly adjusting parameters and verifying outputs, the difference between the machine learning model and the physical model is gradually narrowed to obtain the hydrological risk machine model after parameter adjustment and the hydrological risk physical model after parameter adjustment.

[0047] Step 208, using the adjusted hydrological risk machine model as the hydrological risk machine model, and using the adjusted hydrological risk physical model as the hydrological risk physical model, and returning to execute the step of inputting the real-time hydrological data and the historical hydrological data into the hydrological risk machine model and the hydrological risk physical model respectively, to obtain the risk machine model prediction data and the risk physical model prediction data.

[0048] Specifically, after the parameters of the hydrological risk machine model and the hydrological risk physical model are adjusted respectively, the adjusted hydrological risk machine model is used as the hydrological risk machine model, and the adjusted hydrological risk physical model is used as the hydrological risk physical model, and the hydrological real-time data and hydrological historical data are re-input into the adjusted hydrological risk machine model and the adjusted hydrological risk physical model for new predictions. At this time, the deviation of the output data of the two new models should be reduced. If the difference between the predicted data of the risk machine model and the predicted data of the risk physical model is still greater than the preset threshold, it means that one or both models still need to be further optimized, so the model parameters need to be adjusted further. Through the iterative process of continuously optimizing the model parameters, until the difference between the predicted data of the risk machine model and the predicted data of the risk physical model is less than the preset difference, it means that the prediction models in two different directions can output the same prediction results, reducing the possibility of model errors.

[0049] Step 210, until the difference between the risk machine model prediction data and the risk physical model prediction data is less than the preset difference, select the risk machine model prediction data or the risk physical model prediction data as the water area risk prediction data.

[0050] Among them, the water area risk prediction data can be data selected from one of the water area risk assessment results obtained from the hydrological risk machine model and the hydrological risk physical model.

[0051] Specifically, when the difference between the risk machine model prediction data and the risk physical model prediction data is less than the preset difference, that is, when the difference between the prediction results of the hydrological risk machine model and the hydrological risk physical model is within the tolerance range, it is necessary to select one of the prediction results of the risk machine model prediction data or the risk physical model prediction data as the water area risk prediction data. At this time, it can be decided whether to select the prediction results of the hydrological risk machine model or the hydrological risk physical model based on the actual situation. For example, if a model performs better in a specific water environment, the prediction results of the model can be given priority; or more reliable prediction data can be selected based on the interpretability and credibility of the model.

[0052] Step 212, performing an availability check on the water area risk prediction data, and when the availability check result indicates that the water area risk prediction data is credible, the water area risk prediction data is used as target risk prediction data.

[0053] Among them, the availability check can be a process of verifying the quality of water area risk prediction data, with the aim of ensuring the reliability and applicability of the prediction data.

[0054] Among them, the target risk prediction data may refer to water area risk prediction data that has been checked for availability and verified to be credible, and these data are ultimately used as a basis for decision support.

[0055] Specifically, the availability check includes verification in multiple dimensions: first, check the integrity of the water area risk prediction data to ensure that the water area risk prediction data is not missing and that all indicators can be calculated normally or meet objective conditions; second, evaluate the accuracy of the water area risk prediction data to determine whether the water area risk prediction data is a reasonable objective law constructed by known historical data or reference data, and ensure that there is no obvious deviation or error in the model output; third, check the timeliness of the water area risk prediction data to confirm that the water area risk prediction data is based on the latest hydrological conditions and meteorological forecast results and is applicable to the current risk prediction scenario; finally, by comparing the water area risk prediction data with historical observation data or using a retrospective analysis method, further confirm the credibility of the water area risk prediction data. If the water area risk prediction data passes the inspection in these aspects, it means that the data is reliable and can be used, and can be used as target risk prediction data. If the inspection results show that the water area risk prediction data has defects or does not conform to the actual situation, it is necessary to re-evaluate the accuracy of the model or re-collect data to improve the reliability and practicality of the prediction results.

[0056] In the above-mentioned multi-dimensional hydrological risk assessment method, the real-time hydrological data and historical hydrological data of the target water area are input into the hydrological risk machine model and the hydrological risk physical model. By comparing the prediction results of the two models, when the difference exceeds the preset threshold, the physical parameters in the model will be automatically adjusted to reduce the model error; after multiple adjustments and iterations, the most credible risk prediction data can be finally selected to ensure the effectiveness of the prediction results in practical applications. Further availability checks of water area risk prediction data further enhance the credibility of the model output and avoid decision-making errors caused by model errors or data inconsistencies. Due to the integration of the advantages of machine learning and physical modeling, the scientificity and practicality of water area risk prediction are improved, so it can effectively improve the accuracy and timeliness of hydrological risk prediction results, and provide more accurate and real-time data support for water area management, disaster warning and decision-making.

[0057] In an exemplary embodiment, Figure 3 As shown, the hydrological risk physical model includes a water quality-flow coupling sub-model and a terrain-flood coupling sub-model; real-time hydrological data and historical hydrological data are input into the hydrological risk physical model to obtain risk physical model prediction data, including steps 302 to 306.

[0058] in:

[0059] Step 302: Input the real-time hydrological data and the historical hydrological data into the water quality-flow coupling sub-model to obtain the water quality-flow coupling data.

[0060] Among them, the water quality flow coupling sub-model can be a mathematical model that describes the interaction between water flow and water quality. Its core principle is that the dynamic changes of water flow will affect the diffusion, propagation and sedimentation of pollutants in the water, and the concentration changes of pollutants will affect the characteristics of water flow.

[0061] The water quality flow coupling data can be the data calculated by the water quality flow coupling sub-model, which reflects the relationship between water flow and water quality. These data include the velocity and flow of water flow under specific hydrological conditions and their impact on the diffusion of pollutants in water.

[0062] Specifically, the real-time hydrological data and the historical hydrological data are input into the water quality-flow coupling sub-model. The water quality-flow coupling sub-model mainly focuses on the interaction between water flow and water quality. It estimates water quality changes by simulating the water flow velocity, flow and the diffusion, sedimentation, and degradation of pollutants in the water. For example, the concentration changes of dissolved oxygen, nitrogen and phosphorus pollutants, heavy metals and other components in the river can be calculated. Therefore, the water quality-flow coupling sub-model first calculates the flow velocity, flow and its distribution of the water flow, and uses the water flow equation (such as the Navier-Stokes equation or the shallow water equation) to describe the spatial and temporal changes of the water flow. Then, the diffusion, sedimentation, and degradation of pollutants in the water flow are simulated by the water quality transport equation (such as the Advance-Diffusion equation). During the calculation process, the water quality-flow coupling sub-model will also consider the changes in water quality, such as the concentration changes of pollutants such as dissolved oxygen, nitrogen and phosphorus, and heavy metals in the water. Finally, the water quality-flow coupling sub-model outputs the coupling results of water quality and water flow, thereby obtaining water quality-flow coupling data.

[0063] Among them, the expression of the water quality flow coupling sub-model is:

[0064]

[0065] Among them, Q1(t) is the first water flow at time t, P(t) is the real-time precipitation at time t, Q h (t) is the historical flow at time t, A is the basin area, λ is the coupling adjustment coefficient of precipitation and basin area, δ is the influence coefficient of historical flow on current flow, α is the influence coefficient of historical flow on current flow prediction, κ is the attenuation rate of the influence of historical flow on flow, C(t) is the concentration of water quality factor at time t, D is the diffusion rate of substances inside the water body, V is the water volume, γ is the coefficient of the influence of precipitation on water quality factor at time t, f(·) is the nonlinear influence function of the first water flow at time t, τC(t) is the feedback effect of water quality factor concentration on flow at time t, It is the cumulative impact of the coupled effect of flow and water quality concentration over time.

[0066] Step 304 , inputting the real-time hydrological data and the historical hydrological data into the terrain-flood coupling sub-model to obtain terrain-flood coupling data.

[0067] Among them, the terrain-flood coupling submodel can be a mathematical model that describes the interaction between hydrological conditions and terrain characteristics. This model helps predict the occurrence, spread and impact range of floods by simulating the effects of precipitation, basin terrain (such as slope, riverbed morphology) and other factors on flood formation and propagation.

[0068] Among them, terrain-flood coupling data can be data generated by the terrain-flood coupling sub-model, reflecting information such as flood propagation, waterlogging area, and water level changes under specific terrain conditions and hydrological conditions. These data include the flow path of the flood, the inundated area, the speed of the water flow, and the water level changes.

[0069] Specifically, the real-time hydrological data and historical hydrological data are also input into the terrain-flood coupling sub-model. The terrain-flood coupling sub-model mainly focuses on simulating the interaction between hydrological conditions and terrain characteristics, especially in the flood scenario caused by precipitation, how terrain characteristics affect the propagation path of water flow, the scope of floods and the speed of water flow. The topography of the basin (such as mountains, plains, riverbeds, etc.) determines how precipitation converges into rivers and eventually leads to floods. For example, steep mountains may cause rapid water convergence after heavy rains, resulting in rapid floods, while flat areas may experience water accumulation or stagnation. Therefore, the terrain-flood coupling sub-model uses terrain characteristics and hydrodynamic equations to simulate the propagation process of water flow. The shallow water equation (Saint-Venant equation) or hydrodynamic model is used to describe the changes in water level and flow velocity; according to precipitation and terrain conditions, the model calculates the convergence process of water flow (for example, the convergence of water flow from upstream to downstream) and the distribution path of water flow; by considering different terrain conditions (such as mountains, plains, etc.), the model further calculates the spread, source and flooded area of ​​water flow. On this basis, the model will calculate the spatiotemporal evolution of floods, simulate the water level rise and possible flood impacts in different areas, and couple the above data to obtain terrain-flood coupling data.

[0070] Among them, the expression of the terrain flood coupling sub-model is:

[0071]

[0072] Among them, Q2(t) is the flow rate of the second water flow at time t, h(x,y,t) is the spatial distribution of the water level, v(x,y,t) is the velocity vector, P(t) is the real-time precipitation at time t, S is the water area, F(t) is the flood risk index of any area in the water area at time t, σ(x,y) is the water accumulation coefficient of the basin, θ(x,y,t) is the water flow resistance coefficient, DEM(t) is the real-time terrain data, Flood source function combining precipitation, discharge and topographic effects.

[0073] Step 306 , performing dynamic feedback coupling processing on the water quality-flow coupling data and the terrain-flood coupling data to obtain risk physical model prediction data.

[0074] Among them, dynamic feedback coupling processing can be due to the mutual influence of water quality changes and flood propagation, and the outputs of the two models need to be adjusted according to real-time data to ensure that the model can accurately reflect the interaction between them. In the dynamic feedback process, changes in water flow may affect the spread of pollutants, while changes in pollutant concentrations will affect the dynamic characteristics of water flow, and vice versa.

[0075] Specifically, since the water quality-flow coupling data provides the relationship between the diffusion of pollutants and water flow, and the terrain-flood coupling data provides the propagation path and impact range of floods, the two prediction results will be combined for coupling adjustment. For example, the water flow path and water level changes predicted by the terrain flood model will affect the transmission speed and concentration of pollutants, while water quality changes (such as changes in pollutant concentrations) may change the dynamic characteristics of water flow and affect the spread of floods and water level changes. By introducing nonlinear adjustment factors, especially when the interaction between water quality and floods is strong, it is ensured that the output results of the two models can reflect the complex relationship between water quality changes and flood dynamics, self-optimize, and adjust various parameters such as water flow, pollutant diffusion, and flood propagation, so that the final risk physics model prediction data can more accurately reflect the actual hydrological risk.

[0076] The expression of dynamic feedback coupling processing is:

[0077]

[0078] Among them, R final(t) is the risk physics model prediction data, ρ1 is the hydrological risk feedback coefficient, ρ2 is the flood risk feedback coefficient, ξ1 is the hydrological risk nonlinear adjustment coefficient, and ξ2 is the flood risk nonlinear adjustment coefficient. ρ1 and ρ2 are time-adaptive and can be adjusted through historical data and different scenarios; ξ1 and ξ2 can be set through data fitting or experiments, depending on factors such as watershed, climate, historical floods, etc., and dynamically adjusted based on real-time data.

[0079] In this embodiment, by inputting real-time hydrological data and historical data into the water quality flow coupling sub-model and the terrain flood coupling sub-model, detailed prediction data related to water quality and floods can be obtained respectively. The water quality flow coupling sub-model helps simulate the interaction between water flow and water quality, and evaluates the diffusion, sedimentation and degradation process of pollutants in the water, while the terrain flood coupling sub-model can reflect the interaction between hydrological conditions and terrain, and predict the occurrence and propagation of floods. By subjecting these two types of data to dynamic feedback coupling processing, the mutual influence between water quality and floods can be more accurately simulated, and comprehensive risk physical model prediction data can be obtained. Overall, it can provide more comprehensive and accurate data support for water area risk prediction, help decision makers better deal with pollution prevention and control and flood management in water areas, and improve the reliability of risk assessment and the scientific nature of decision-making.

[0080] In an exemplary embodiment, Figure 4 As shown, the water area risk prediction data is checked for availability, and when the availability check result indicates that the water area risk prediction data is credible, the water area risk prediction data is used as the target risk prediction data, including steps 402 to 408.

[0081] Step 402, construct risk data constraint conditions based on real-time hydrological data, historical hydrological data and real-time terrain data of the target water area.

[0082] Real-time terrain data can be detailed information about the target water area and its surrounding terrain obtained through remote sensing technology, geographic information system (GIS) or other monitoring tools within a specific time. Such data include but are not limited to the terrain undulations, slope, riverbed morphology, watershed distribution, soil type and other geographical factors that may affect water flow and water level.

[0083] Among them, risk data constraints can be restrictions or rules set by actual observation data, historical hydrological data, real-time terrain data, etc. when predicting water area risks. These conditions are used to ensure that the predicted data output by the hydrological model conforms to the actual situation and avoid unreasonable results. For example, water level data may be restricted by certain physical or geographical features, the flow rate cannot exceed a certain critical value, or under certain terrain conditions, the water flow in certain areas cannot reach a certain preset range.

[0084] Specifically, when constructing risk data constraints for water area risk prediction data, it is necessary to make full use of the real-time hydrological data, historical hydrological data, and real-time terrain data of the current water area to determine reasonable boundary conditions. These data are not limited to current water level, flow rate, precipitation, historical basin flood events, and terrain information of the water area (such as slope, riverbed morphology, watershed area, etc.). By integrating these factors, constraints on water area risk prediction data can be established to ensure that all prediction results are within the physically and hydrologically feasible range. For example, the water level should not exceed a certain height, and the flow rate should meet certain physical restrictions. The establishment of these constraints can effectively limit the output range of the model and avoid prediction results that are not in line with reality.

[0085] Step 404, performing a hydrological logic check on the water area risk prediction data according to the risk data constraint conditions to obtain risk data logic check data.

[0086] Among them, the hydrological logic check can be a verification of the water area risk prediction data, which mainly checks whether the hydrological data output by the model conforms to the actual hydrological laws and physical logic.

[0087] The risk data logic check data may be the inspection result after the hydrological logic check, which is used to further ensure that there are no logical problems with the prediction data.

[0088] Specifically, according to the risk data constraints, by checking the hydrological parameters such as water level, flow rate, precipitation, etc. in the water area risk prediction data, ensure that the relationship between them is reasonable, and obtain the risk data logic check data. For example, the change of water level over time should be stable, and there should be no sudden extreme fluctuations; there should be a certain physical relationship between water flow and water level, and there should be no abnormal situation of excessive flow or too low water level. In addition, the hydrological logic check process will also verify whether the model output is consistent with the actual hydrological law by comparing with the actual historical observation results. If there is illogical situation in the predicted data, such as unreasonable fluctuations in water level or abnormal changes in flow rate, the system will mark it and perform further processing.

[0089] Step 406, when the risk data logic check data indicates that the water area risk prediction data does not have hydrological logic anomalies, a credibility analysis is performed on the water area risk prediction data to obtain a risk data credibility analysis result.

[0090] Among them, credibility analysis can be an evaluation process of water risk prediction data, the purpose of which is to verify the reliability and accuracy of the prediction results. By back-analyzing and verifying the performance of the model in similar situations, credibility analysis can evaluate the accuracy of the model output. The analysis can also include error analysis to detect the difference between the model prediction data and the actual data. If the prediction results are consistent with the historical data and the error is within an acceptable range, then the prediction data can be considered credible.

[0091] Among them, the results of the credibility analysis of risk data can be the conclusions drawn after the credibility analysis of the water area risk prediction data. The results indicate the reliability of the prediction data, which is specifically manifested in the error between the prediction result and the actual observation data, the error range of the model, and the degree of conformity of the prediction data. If the credibility analysis results show that the prediction data error is small and consistent with historical data and physical constraints, then these data are considered credible. If the credibility analysis results show that the data has large deviations or inconsistencies, it is necessary to readjust the model or further correct the input data.

[0092] Specifically, when the risk data logic check data characterizes that there is no hydrological logic anomaly in the water area risk prediction data, the next step is to conduct a credibility analysis on the water area risk prediction data. The credibility analysis mainly verifies the accuracy of the model output results by comparing the prediction data with the historical observation data. It includes retrospective analysis, that is, using the model to predict historical events and comparing them with the actual results to check whether the model can accurately predict risks in similar historical situations; in addition, the credibility analysis also includes error assessment, checking the error range between the model output and the actual observation value. For the credibility of the prediction data, an error tolerance threshold is usually set. Only when the error of the prediction result is within the tolerance range, the data is considered credible, ensuring that the final prediction data has sufficient reliability and obtaining the credibility analysis results of the risk data.

[0093] Step 408: When the risk data credibility analysis result indicates that the water area risk prediction data is credible, the water area risk prediction data is used as target risk prediction data.

[0094] Specifically, when the results of the risk data credibility analysis characterize that the water area risk prediction data is credible, that is, the credibility analysis shows that the water area risk prediction data meets the accuracy requirements and there is no obvious deviation or non-conformity with reality, the final prediction data will be deemed credible and returned as the target risk prediction data; if the credibility analysis fails, it is necessary to readjust the model or further correct the input data to ensure that the prediction data is consistent with the actual situation and the actual risk needs of the water area, so as to ensure the effectiveness and operability of the decision.

[0095] In this embodiment, by constructing risk data constraints based on real-time hydrological data, historical data, and real-time terrain data of the target water area, it is possible to ensure that the water area risk prediction data is within the scope of physical and hydrological laws, avoiding unreasonable prediction results. Through hydrological logic checks, it is further verified whether the risk prediction data conforms to hydrological laws to ensure that there are no logical problems such as abnormal water levels and unreasonable flow rates. If the risk data logic check passes, a credibility analysis is performed to ensure the reliability of the prediction results through historical comparison of the data and model verification. Finally, when the credibility analysis results show that the prediction data is credible, it can be used as the target risk prediction data for actual decision-making, ensuring the scientificity, accuracy and operability of the risk prediction data, providing reliable data support for water area risk management and emergency response, and effectively improving the quality and efficiency of decision-making.

[0096] In an exemplary embodiment, Figure 5 As shown, the credibility analysis of the water area risk prediction data is performed to obtain the risk data credibility analysis result, including steps 502 to 506. Among them:

[0097] Step 502, using Latin hypercube sampling, quantify the uncertainty of water area risk prediction data to obtain risk data uncertainty information.

[0098] Among them, risk data uncertainty information can be the uncertainty level of water area risk prediction data obtained by analyzing the range and uncertainty of model input parameters. These uncertainties usually come from factors such as changes in input data, measurement errors, model assumptions, etc. For example, input data such as precipitation, flow rate, and topography may have certain prediction errors or fluctuations.

[0099] Specifically, since Latin Hypercube Sampling (LHS) is a statistical sampling method, it is mainly used to quantify uncertainty and improve the representativeness of simulation results. Therefore, the input variables of the risk prediction model (such as water level, flow velocity, precipitation, etc.) are first determined, and then the LHS method is used to uniformly extract samples from the possible range of water area risk prediction data to ensure that the sampling results cover all possible scenarios. Multiple groups of different input combinations can be generated and model simulations can be performed separately to obtain a series of water area risk prediction data. These prediction data are used to quantify the uncertainty of the prediction results and reveal the changes in the prediction results under different input conditions. Finally, the multiple groups of results generated by LHS sampling can be used to calculate the uncertainty information of the water area risk prediction data.

[0100] Step 504: Perform sensitivity analysis on the water area risk prediction data to obtain risk data sensitivity information.

[0101] Among them, sensitivity analysis can be a method to evaluate the influence of input variables on model output results. In water area risk prediction, sensitivity analysis identifies which factors play an important role in model output by systematically changing input variables (such as water level, flow rate, precipitation, etc.) and observing their impact on the results (such as flood risk, water level change, etc.).

[0102] Among them, the risk data sensitivity information can be obtained through sensitivity analysis, which reflects the specific degree of influence of different input variables on the water area risk prediction results. This information reveals which factors have a significant impact on the output data of the model. For example, if an input variable (such as precipitation or flow rate) changes, and the model prediction results change significantly, it means that the variable has a high sensitivity to the prediction results.

[0103] Specifically, we first select the key input variables that affect the water area risk prediction, such as the precipitation, flow rate, and terrain characteristics of the basin. Then, we conduct systematic tests on the changes in these input variables and analyze their impact on the water area risk prediction data (such as water level, flow, etc.). In the process of sensitivity analysis, we usually identify which factors have a greater impact on the water area risk prediction data by gradually changing the values ​​of the input variables and observing the changes in the output of the water area risk prediction data. Through sensitivity analysis, we can determine which input variables are the most sensitive and which factors' changes will significantly affect the risk prediction results, and obtain the sensitivity information of the risk data.

[0104] Step 506: Determine the risk data credibility analysis result according to the risk data uncertainty information and the risk data sensitivity information.

[0105] Specifically, since the risk data uncertainty information reveals the possible fluctuation range of different input variables during the simulation process, these fluctuations reflect the sensitivity of the model to various uncertainties. The risk data sensitivity information helps identify which variables play a decisive role in the changes in risk prediction results by evaluating the degree of influence of the input variables on the prediction results. When combining the two, the risk data uncertainty information is used to determine which key input variables have a larger fluctuation range, thereby affecting the stability of the results; at the same time, the risk data sensitivity information can indicate which variables have a significant impact on the model output. Finally, by taking this information into consideration, the performance and reliability of the model under different conditions are evaluated, and a comprehensive risk data credibility analysis result is obtained.

[0106] In this embodiment, by using Latin hypercube sampling to quantify the uncertainty of water area risk prediction data, the potential fluctuations in input data and model predictions can be comprehensively evaluated, thereby revealing the range of variation and sources of uncertainty in the prediction results. Then, by performing a sensitivity analysis on the risk prediction data, it is possible to identify which input variables have the greatest impact on the model results, helping to determine the key factors that need to be focused on in the model. Combining uncertainty information and sensitivity information to conduct a credibility analysis of risk data can not only quantify the contribution of different factors to the results, but also comprehensively evaluate the reliability of the prediction data. Overall, it improves the transparency and credibility of risk predictions, provides more accurate and secure data support for water area risk management, and helps decision makers make more effective responses.

[0107] In an exemplary embodiment, Figure 6 As shown, the risk data credibility analysis result is determined according to the risk data uncertainty information and the risk data sensitivity information, including steps 602 to 606.

[0108] in:

[0109] Step 602: perform abnormal inference on the risk data uncertainty information and the risk data sensitivity information according to the abnormal detection algorithm to obtain abnormal inference data.

[0110] Among them, anomaly detection algorithms can be technologies used to identify data points in a data set that do not conform to expected patterns or deviate from conventional behavior. In water risk prediction, anomaly detection algorithms are used to discover data anomalies that may be caused by measurement errors, inaccurate model assumptions, or extreme events. Common anomaly detection methods include statistical methods (such as standard deviation method, box plot method) and machine learning methods (such as isolation forest, support vector machine, etc.).

[0111] Among them, anomaly inference can be a further analysis process of data anomalies discovered by anomaly detection algorithms. Anomaly inference infers potential anomaly causes or abnormal patterns by deeply mining the data.

[0112] Among them, abnormal inference data can be the result of the abnormal inference process, which represents the abnormal patterns or data points identified in the data set. These data may be data values ​​that deviate from the norm in the prediction process, or changes caused by external factors (such as extreme climate events, sudden disasters, etc.).

[0113] Specifically, anomaly detection algorithms are used to analyze the uncertainty and sensitivity information of risk data to find parts of the data that do not conform to normal patterns. For example, by performing statistical analysis on input variables (such as water level, flow rate, precipitation, etc.), data points that are far from the normal range are identified. These data points may be caused by abnormal input, incorrect measurement or model assumptions, and outliers in the data are identified to obtain abnormal inference data.

[0114] Step 604 , performing deviation propagation analysis on the risk data uncertainty information and the risk data sensitivity information according to the deviation propagation model to obtain deviation propagation analysis data.

[0115] Among them, the bias propagation model can be a mathematical model that evaluates how the uncertainty of input variables propagates in the model calculation and ultimately affects the model output. In water risk prediction, the bias propagation model propagates the uncertainty of input data (such as measurement error, estimation bias, etc.) to each level of the model to evaluate the impact of these uncertainties on the final prediction results. The model calculates how the deviation of input data propagates during the calculation process through mathematical formulas (such as Taylor expansion, Monte Carlo method, etc.), and quantifies the impact of this propagation on the prediction results, thereby helping to understand the sources of uncertainty and the range of variation in the prediction.

[0116] Among them, bias propagation analysis can be an analysis process performed by the bias propagation model to quantify how the bias of the input variables affects the final prediction results, and to evaluate their contribution to the model output by systematically analyzing the uncertainty of each input variable.

[0117] Among them, the deviation propagation analysis data can be the results obtained from the deviation propagation analysis process, indicating which factors have a greater impact on the final prediction results during the deviation propagation process of the input variables. These data reveal which input variables in the model have significantly contributed to the changes in the risk prediction results, helping to identify which factors need to be focused on and controlled.

[0118] Specifically, risk data uncertainty information provides the uncertainty range of input variables (such as precipitation, flow rate, water level, etc.), which usually come from measurement errors, model assumptions or natural changes. The risk data sensitivity information provides the degree of influence of each input variable on the prediction results, that is, identifying which variables have a higher sensitivity to the final results. For example, small changes in certain variables may cause large fluctuations in the prediction results, while changes in other variables have little effect on the results. Based on this uncertainty and sensitivity information, the deviation propagation model propagates the deviations of the input variables to the output layer of the model, calculates the specific contribution of these deviations to the final result, quantifies how the uncertainty of each input parameter accumulates during the prediction process, and identifies which factors play a dominant role in the model. Finally, the deviation propagation analysis data is obtained.

[0119] Step 606, multiplying the abnormal inference data with the deviation propagation analysis data to obtain the risk data credibility analysis result.

[0120] Specifically, the anomaly inference data is multiplied with the deviation propagation analysis data to comprehensively consider the impact of abnormal information and deviation propagation on the credibility of risk data; the anomaly inference data provides clues about which data points may have anomalies, while the deviation propagation analysis data reveals the degree of influence of different sources of uncertainty on the results. By multiplying the anomaly inference data with the deviation propagation analysis data, a comprehensive risk data credibility analysis result can be obtained, which reflects the uncertainty of the data and the degree of influence of the anomaly, and can provide a quantitative evaluation of the accuracy of water risk prediction.

[0121] In this embodiment, by using an anomaly detection algorithm to perform anomaly inference on the uncertainty information and sensitivity information of the risk data, abnormal data that deviates from the normal pattern can be effectively identified and eliminated, ensuring that the analysis results are more accurate and reliable. Further using the deviation propagation model to perform deviation propagation analysis on uncertainty and sensitivity information can quantify how the uncertainty of the input data affects the final prediction results, thereby helping to identify and reduce the propagation of errors. By combining the anomaly inference data with the deviation propagation analysis data, it is possible to comprehensively consider the abnormal factors and deviation effects in the data to obtain a more comprehensive and accurate risk data credibility analysis result. Overall, it improves the scientificity and credibility of risk prediction, provides stronger support for water risk assessment and decision-making, and helps to formulate more accurate management and response measures.

[0122] Based on the same inventive concept, the embodiment of the present application also provides a multi-dimensional hydrological risk assessment device for implementing the multi-dimensional hydrological risk assessment method involved above, such as Figure 7As shown, it includes: a hydrological data acquisition module 702, a hydrological data analysis module 704, a hydrological model adjustment module 706, a prediction data selection module 708 and a risk data determination module 710. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations of the one or more multi-dimensional hydrological risk assessment device embodiments provided can refer to the above limitations on a multi-dimensional hydrological risk assessment method, which will not be repeated here.

[0123] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface; wherein the memory stores a computer program, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0124] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0125] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0126] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0127] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A multi-dimensional hydrological risk assessment method, characterized in that: The method comprises: Obtain real-time and historical hydrological data of the target waters; Inputting the real-time hydrological data and the historical hydrological data into the hydrological risk machine model and the hydrological risk physical model respectively, to obtain risk machine model prediction data and risk physical model prediction data; When the difference between the risk machine model prediction data and the risk physical model prediction data is greater than a preset difference, adjusting the model parameters of the hydrological risk machine model and the hydrological risk physical model according to the difference; The adjusted hydrological risk machine model is used as the hydrological risk machine model, and the adjusted hydrological risk physical model is used as the hydrological risk physical model, and the step of inputting the hydrological real-time data and the hydrological historical data into the hydrological risk machine model and the hydrological risk physical model respectively to obtain the risk machine model prediction data and the risk physical model prediction data is returned to execute; Until the difference between the risk machine model prediction data and the risk physical model prediction data is less than the preset difference, the risk machine model prediction data or the risk physical model prediction data is selected as the water area risk prediction data; An availability check is performed on the water area risk prediction data, and when a check result of the availability check indicates that the water area risk prediction data is credible, the water area risk prediction data is used as target risk prediction data.

2. The method according to claim 1, characterized in that The hydrological risk physical model includes a water quality-flow coupling submodel and a terrain-flood coupling submodel; The step of inputting the real-time hydrological data and the historical hydrological data into the hydrological risk physical model to obtain risk physical model prediction data includes: Inputting the real-time hydrological data and the historical hydrological data into the water quality-flow coupling sub-model to obtain water quality-flow coupling data; Inputting the real-time hydrological data and the historical hydrological data into the terrain-flood coupling sub-model to obtain terrain-flood coupling data; Dynamic feedback coupling processing is performed on the water quality flow coupling data and the terrain flood coupling data to obtain the risk physical model prediction data.

3. The method according to claim 2, characterized in that The expression of the water quality flow coupling sub-model is: Among them, Q1(t) is the first water flow at time t, P(t) is the real-time precipitation at time t, Q h (t) is the historical flow at time t, A is the basin area, λ is the coupling adjustment coefficient of precipitation and basin area, δ is the influence coefficient of historical flow on current flow, α is the influence coefficient of historical flow on current flow prediction, k is the attenuation rate of the influence of historical flow on flow, C(t) is the concentration of water quality factor at time t, D is the diffusion rate of substances inside the water body, V is the water volume, γ is the coefficient of the influence of precipitation on water quality factor at time t, f(·) is the nonlinear influence function of the first water flow at time t, τC(t) is the feedback effect of water quality factor concentration on flow at time t, It is the cumulative impact of the coupled effect of flow and water quality concentration over time.

4. The method according to claim 2, characterized in that: The expression of the terrain flood coupling submodel is: Among them, Q2(t) is the flow rate of the second water flow at time t, h(x,y,t) is the spatial distribution of the water level, v(x,y,t) is the velocity vector, P(t) is the real-time precipitation at time t, S is the water area, F(t) is the flood risk index of any area in the water area at time t, σ(x,y) is the water accumulation coefficient of the basin, θ(x,y,t) is the water flow resistance coefficient, DEM(t) is the real-time terrain data, Flood source function combining precipitation, discharge and topographic effects.

5. The method according to claim 2, characterized in that: The expression of the dynamic feedback coupling process is: Among them, R final(t) is the prediction data of the risk physics model, ρ1 is the hydrological risk feedback coefficient, ρ2 is the flood risk feedback coefficient, ξ1 is the hydrological risk nonlinear adjustment coefficient, and ξ2 is the flood risk nonlinear adjustment coefficient.

6. The method according to claim 1, characterized in that The performing of an availability check on the water area risk prediction data, and using the water area risk prediction data as target risk prediction data when the check result of the availability check indicates that the water area risk prediction data is credible, comprises: Constructing risk data constraint conditions according to the real-time hydrological data, the historical hydrological data and the real-time terrain data of the target water area; According to the risk data constraint conditions, a hydrological logic check is performed on the water area risk prediction data to obtain risk data logic check data; When the risk data logic check data indicates that the water area risk prediction data does not have hydrological logic anomalies, a credibility analysis is performed on the water area risk prediction data to obtain a risk data credibility analysis result; When the risk data credibility analysis result indicates that the water area risk prediction data is credible, the water area risk prediction data is used as the target risk prediction data.

7. The method according to claim 6, characterized in that The credibility analysis of the water area risk prediction data is performed to obtain the risk data credibility analysis result, including: Use Latin hypercube sampling to quantify the uncertainty of the risk prediction data of the water area and obtain the uncertainty information of the risk data; Performing sensitivity analysis on the water area risk prediction data to obtain risk data sensitivity information; The risk data credibility analysis result is determined according to the risk data uncertainty information and the risk data sensitivity information.

8. The method according to claim 7, characterized in that The step of determining the risk data credibility analysis result according to the risk data uncertainty information and the risk data sensitivity information includes: According to the anomaly detection algorithm, abnormal inference is performed on the risk data uncertainty information and the risk data sensitivity information to obtain abnormal inference data; According to the deviation propagation model, performing deviation propagation analysis on the risk data uncertainty information and the risk data sensitivity information to obtain deviation propagation analysis data; The abnormal inference data is multiplied by the deviation propagation analysis data to obtain the risk data credibility analysis result.

9. A multi-dimensional hydrological risk assessment device, characterized in that: The device comprises: A hydrological data acquisition module is used to obtain real-time and historical hydrological data of the target water area; A hydrological data analysis module, used to input the real-time hydrological data and the historical hydrological data into a hydrological risk machine model and a hydrological risk physical model, respectively, to obtain risk machine model prediction data and risk physical model prediction data; A hydrological model adjustment module, configured to adjust the model parameters of the hydrological risk machine model and the hydrological risk physical model according to the difference between the risk machine model prediction data and the risk physical model prediction data when the difference is greater than a preset difference; The hydrological data analysis module is further used to use the adjusted hydrological risk machine model as the hydrological risk machine model, and the adjusted hydrological risk physical model as the hydrological risk physical model, and return to execute the step of inputting the hydrological real-time data and the hydrological historical data into the hydrological risk machine model and the hydrological risk physical model respectively to obtain the risk machine model prediction data and the risk physical model prediction data; A prediction data selection module, used for selecting the risk machine model prediction data or the risk physical model prediction data as the water area risk prediction data when the difference between the risk machine model prediction data and the risk physical model prediction data is less than the preset difference; The risk data determination module is used to perform an availability check on the water area risk prediction data, and when the check result of the availability check indicates that the water area risk prediction data is credible, use the water area risk prediction data as target risk prediction data.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images