River risk simulation treatment scheme optimization system based on scheme rehearsal

Abstract

The invention discloses a riverway risk simulation treatment scheme optimization system based on scheme rehearsal, and relates to the field of data simulation, and the system comprises a management module which is used for obtaining the access authority of each riverway and hydrological data of a target water area, and providing the editing authority of control instruction data flow of each function module; the riverway simulation unit is used for dividing a target riverway into a plurality of river sections for virtual construction of a three-dimensional model according to the riverway and hydrological data of an actual water area, simulating a water flow rate, a water level, a submerging range and a water depth based on the hydrological data, and simulating a riverbed sediment state and river levee strength data based on the riverway data; by performing dynamic simulation and correlation analysis on multiple schemes in a virtual digital twinning environment, potential risks and side effects of the schemes in long-term operation and extreme scenes can be foreseen in advance, and multiple pressure tests and trial and error optimization are completed before work, so that a real and reliable scheme is screened out, and the working efficiency is improved. And the risk of engineering failure or poor effect is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of data simulation technology, specifically to a system for optimizing river risk simulation and management schemes based on scheme pre-simulation. Background Technology

[0002] Climate change is altering rainfall patterns, and the increasing frequency and intensity of extreme rainfall events are severely testing the reliability and safety of traditional river management projects designed based on historical hydrological data. Modern river management no longer focuses solely on flood control but requires a holistic approach that considers water resource security, ecological health, and socio-economic development. This necessitates a successful solution that balances flood control with ecological restoration, landscape enhancement, the creation of waterfront spaces, and cost control, making the decision-making process exceptionally complex and requiring the balancing of multiple interests and sometimes conflicting objectives.

[0003] Traditional methods rely heavily on engineers' personal experience and historical cases. Design schemes are usually based on a fixed design standard, making it difficult to go beyond the standard design scenario to analyze the chain reaction that the scheme will have on downstream river sections after implementation. As a result, ecological risks and long-term structural instability risks are easily overlooked, and it is difficult to consider the correlation and transmission relationship between different risks. Potential defects in the governance scheme are often only discovered after the project is implemented, or even after a catastrophic event, and the cost of correction is huge. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a river risk simulation and management scheme optimization system based on scheme pre-simulation, which can effectively solve the problems of the existing technology.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] This invention discloses a system for optimizing river risk simulation and management schemes based on scheme pre-simulation, comprising:

[0009] The management module is used to obtain access permissions for various river and hydrological data of the target water area, and provides editing permissions for the control command data stream of each functional module;

[0010] The river simulation unit is used to divide the target river into several sections and virtually construct a three-dimensional model based on the actual waterway and hydrological data. It simulates water flow velocity, water level, inundation range and water depth based on hydrological data, and simulates riverbed sediment state and embankment strength data based on river channel data.

[0011] The event generation module is used to automatically identify and generate several potential risk events corresponding to each river segment based on the simulation results of hydrological and structural safety of each river segment in the current simulation cycle. The risk events include at least dike breach, bank slope instability, excessive scouring or silting of the riverbed, and ecological degradation, and their trigger thresholds and key parameters can be customized by the user.

[0012] The scheme management module is used to acquire and store preset treatment schemes, including the treatment measures and engineering parameters to be taken for one or more risk events in a target river section.

[0013] The indicator extraction module is used to quantify and extract several impact indicators of the governance schemes preset by the scheme management module on various risk events, and simultaneously evaluate the implementation difficulty indicators of the schemes; the impact indicators include engineering cost, expected effect intensity, and ecological disturbance degree, and the implementation difficulty indicators include technical complexity and construction period.

[0014] The intervention prediction module is used to input all the impact indicators and implementation difficulty indicators output by the indicator extraction module into the risk intervention model preset by machine learning. The model jointly predicts the intervention coefficient of each indicator on the corresponding risk event in the future period. The intervention coefficient represents the ability of the governance plan to suppress the probability and severity of the risk event.

[0015] The judgment module is used to compare the future interference coefficients output by the interference prediction module with the preset standard thresholds. When the interference coefficient of any risk event fails to meet its standard threshold, the current governance plan is judged to be substandard.

[0016] The scheme adjustment module is used to construct a correlation state network between various influencing indicators based on historical data when the scheme fails to meet the standards. The correlation state includes the mutual promotion, inhibition or coupling relationship between the indicators. Based on the correlation state network, with the goal of maximizing the total interference coefficient of the system, the optimal adjustment range of the influencing indicators that fail to meet the standards is solved, and the adjusted governance scheme is generated as an optimization reference.

[0017] Furthermore, the river channel simulation unit is further equipped with sub-modules, including a river segment construction module, a hydrodynamic simulation module, and a riverbed simulation module. The river segment construction module is interconnected with the hydrodynamic simulation module and the riverbed simulation module via a wireless network.

[0018] The river segment construction module is used to divide the river into several continuous river segments based on the actual geographic and hydrological data of the target river, and to build a digital twin model for each river segment that includes three-dimensional topography, riverbed quality, levee structure and hydraulic roughness parameters.

[0019] The hydrodynamic simulation module is used to import boundary conditions from hydrological data, run the hydrodynamic model on the digital twin model, and simulate the spatial distribution and temporal evolution of water flow velocity, water level, inundation range and water depth in each river section.

[0020] The riverbed simulation module is used to simulate the sediment erosion and deposition process of the riverbed based on the hydrodynamic simulation results of the hydrodynamic simulation module. It defines the river section parameters of sediment erosion, transport and deposition flux in each river section during the simulation period, and evaluates the structural stability coefficient of the current riverbank parameters under different hydrological conditions.

[0021] Furthermore, the actual geographic and hydrological data in the river section construction module includes lidar point clouds, UAV aerial survey images, multibeam underwater topographic survey data, historical hydrological observation data, and geological exploration data.

[0022] Furthermore, the process of constructing the hydrodynamic model in the hydrodynamic simulation module is as follows:

[0023] Based on the river section construction module, three-dimensional terrain data and river cross-section data are obtained, and grids are generated to form a grid of two-dimensional computing units. Each computing unit is assigned an initial water level and an initial flow velocity.

[0024] The flow process curve at the upstream inlet section and the water level-flow relationship curve at the downstream outlet section are used as the boundary conditions of the model, and the lateral inflow or rainfall intensity distribution is set according to the simulation requirements.

[0025] The governing equations are constructed based on the Saint-Venant equations, and the water level and flow velocity of each computational unit are solved simultaneously in each computational time step using the finite difference method.

[0026] The calculation is performed iteratively according to the time step, simulating the spatiotemporal evolution of hydraulic elements in the entire target river channel, and outputting real-time hydrodynamic data including the vector distribution of water flow velocity, water level elevation, dynamic inundation range and water depth of each river section.

[0027] The hydrodynamic data from the entire field is transmitted in real time to the riverbed simulation module and the event generation module through a standardized data interface, serving as their driving data.

[0028] Furthermore, the process by which the riverbed simulation module evaluates the structural stability coefficient is as follows:

[0029] Receive the output results of the full-field hydrodynamic data of each river section from the hydrodynamic simulation module, as well as the riverbed and embankment status data;

[0030] Based on the limit equilibrium method, the anti-sliding stability analysis of the dike slope is carried out to evaluate its anti-sliding safety factor under adverse conditions such as flood soaking and sudden drop in water level.

[0031] Conduct seepage stability analysis of the embankment and foundation, and assess the seepage gradient;

[0032] For revetment and retaining wall structures, assess their resistance to overturning and sliding.

[0033] Based on the scouring effect of water flow on the toe of the dike and the bank slope, and combined with the scouring and deposition depth simulated by riverbed evolution, the stability of the foundation is assessed.

[0034] Furthermore, the process by which the event generation module generates risk events is as follows:

[0035] Based on the simulation results received by the river simulation unit, key assessment parameters corresponding to each risk event are extracted, including:

[0036] For the risk of dike breach, key assessment parameters include the difference between the water level and the top elevation of the dike, and the dike stability coefficient.

[0037] For the risk of bank slope instability, key assessment parameters include the slope safety factor and the rate of water level change;

[0038] For the risk of excessive scouring or silting of the riverbed, key assessment parameters include the depth of scouring or the thickness of siltation.

[0039] For the risk of ecological degradation, key assessment parameters include water flow velocity, water depth, and duration of inundation.

[0040] The extracted key evaluation parameters are compared with user-defined trigger thresholds. The trigger thresholds are set separately for each type of risk event, and the trigger conditions for exceeding or falling below the thresholds are defined according to the nature of the parameters.

[0041] When any key assessment parameter meets its triggering condition, the corresponding risk event identifier is automatically generated, and a list of potential risk events for that river segment during the current simulation period is summarized and output.

[0042] Furthermore, the construction process of the risk intervention model in the intervention prediction module is as follows:

[0043] Collect and organize a dataset of historical governance schemes. Each sample data includes: an input feature vector: quantitative values ​​of various impact indicators and implementation difficulty indicators included in the sample governance scheme; and an output label vector: the actual intervention coefficient for various risk events obtained through actual monitoring or high-precision post-event simulation after the implementation of the sample governance scheme. The actual intervention coefficient is comprehensively quantified by the percentage reduction in the probability of the risk event and the percentage reduction in the expected loss.

[0044] The input feature vectors are standardized, and feature selection is performed based on feature importance analysis. The random forest algorithm is selected as the base model.

[0045] The constructed dataset is divided into a training set and a test set according to a preset ratio. The selected algorithm is trained using the training set with the goal of minimizing the error between the predicted interference coefficient and the actual interference coefficient. The trained model is then validated using the test set.

[0046] The validated risk intervention model is deployed in the intervention prediction module to predict the intervention coefficient for new input governance schemes.

[0047] Furthermore, the formula for calculating the interference coefficient of the risk interference model in the interference prediction module for any risk event is expressed as follows:

[0048] ;

[0049] In the formula, Represents risk events The future interference coefficient characterizes the ability of the governance scheme to suppress the probability of the occurrence of this risk event. This represents the logistic function, used to map linear combinations to probability values. Represents risk events The model bias term is learned through training data. This represents the total number of influencing indicators. Representing the Each indicator for risk events The weighting coefficients, Representing the Standardized values ​​for indicators that affect performance or are difficult to implement. Representing the The first indicator and the first The interaction weighting coefficients between the indicators are used to capture the joint effect among them and are learned through training data. Representing the The standardized values ​​of the influencing indicators, Representing the The standardized value of each implementation difficulty indicator.

[0050] Furthermore, the working logic of the scheme adjustment module is as follows:

[0051] Based on the current values ​​of all influencing indicators, an adjustable search space is defined for each indicator, thereby initializing a population containing multiple potential adjustment schemes;

[0052] Entering the iterative optimization loop, in each generation, the adjustment plan represented by each individual in the population is substituted into the risk intervention model, the intervention coefficient of all risk events is re-predicted, and the comprehensive implementation difficulty corresponding to each plan is calculated.

[0053] Based on the overall implementation difficulty, assess the suitability of each adjustment plan;

[0054] Individuals in the population are sorted and selected based on the fitness of each adjustment scheme, and the preferred schemes with fitness higher than a preset threshold are retained. The preferred schemes are then subjected to crossover and mutation operations that simulate biological genetics.

[0055] Repeat the above selection, crossover and mutation steps until the preset number of iterations or fitness convergence criteria are reached.

[0056] All non-dominated solutions that satisfy the constraints are selected from the last generation of the population to form a Pareto optimal solution set. Each solution in this set represents the optimal adjustment range of a set of non-compliant impact indicators, serving as a reference for multiple alternative optimizations of the current governance scheme.

[0057] Furthermore, the management module is interconnected with the river simulation unit, the event generation module, and the scheme management module via a wireless network; the index extraction module is interconnected with the event generation module, the scheme management module, and the interference prediction module via a wireless network; and the judgment module is interconnected with the interference prediction module and the scheme adjustment module via a wireless network.

[0058] (III) Beneficial Effects

[0059] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0060] 1. By dynamically simulating and analyzing multiple solutions in a virtual digital twin environment, the potential risks and side effects of the solutions under long-term operation and extreme scenarios can be predicted in advance. Before construction begins, multiple stress tests and trial-and-error optimizations have been completed, thereby selecting truly reliable solutions and greatly reducing the risk of project failure or poor results.

[0061] 2. By establishing a multi-dimensional evaluation index system that includes safety, economy, ecology and society, and by using a multi-objective optimization algorithm, it can automatically search for and present a series of solutions that achieve the best balance among multiple objectives. This allows decision-makers to clearly understand the conflicts and trade-offs between different objectives, thereby breaking down professional barriers, avoiding the creation of other new problems in order to solve a single problem, and ultimately selecting the optimal solution.

[0062] 3. By automatically analyzing the correlation between various influencing indicators when a solution fails to meet the standards, and quickly generating specific optimization and adjustment directions, the original time-consuming and labor-intensive engineering design, which relied on repeated manual adjustments, is transformed into a highly efficient and automated process. This greatly improves the efficiency of planning and design and avoids huge economic losses and negative impacts caused by discovering solution defects only after implementation. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic diagram of the overall framework of the present invention.

[0065] The labels in the diagram represent: 1. Management module; 2. River channel simulation unit; 21. River segment construction module; 22. Hydrodynamic simulation module; 23. Riverbed simulation module; 3. Event generation module; 4. Scheme management module; 5. Indicator extraction module; 6. Interference prediction module; 7. Judgment module; 8. Scheme adjustment module. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0067] The present invention will be further described below with reference to embodiments.

[0068] This embodiment presents a river risk simulation and management scheme optimization system based on scheme pre-drafting, such as... Figure 1 As shown, it includes:

[0069] Management Module 1 is used to obtain access permissions for various river and hydrological data of the target water area, and provides editing permissions for the control command data stream of each functional module.

[0070] River channel simulation unit 2 is used to virtually construct a 3D model of the target river channel by dividing it into several river segments based on actual waterway and hydrological data. It simulates water flow velocity, water level, inundation range, and water depth based on hydrological data, and simulates riverbed sediment conditions and levee strength data based on river channel data. River channel simulation unit 2 has sub-modules, including a river segment construction module 21, a hydrodynamic simulation module 22, and a riverbed simulation module 23. The river segment construction module 21 is interconnected with the hydrodynamic simulation module 22 and the riverbed simulation module 23 via a wireless network.

[0071] The river segment construction module 21 is used to divide the river into several continuous river segments based on the actual geographic and hydrological data of the target river, and to construct a digital twin model for each river segment, including three-dimensional topography, riverbed quality, levee structure and hydraulic roughness parameters; the actual geographic and hydrological data include lidar point clouds, UAV aerial survey images, multibeam underwater topographic survey data, historical hydrological observation data and geological exploration data;

[0072] The hydrodynamic simulation module 22 is used to import boundary conditions from hydrological data, run the hydrodynamic model on the digital twin model, and simulate the spatial distribution and temporal evolution of water flow velocity, water level, inundation range, and water depth in each river segment. The construction process of the hydrodynamic model is as follows:

[0073] Based on the river section construction module 21, three-dimensional terrain data and river cross-section data are acquired, and grids are generated to form a grid of two-dimensional computing units. Each computing unit is assigned an initial water level and an initial flow velocity.

[0074] The flow process curve at the upstream inlet section and the water level-flow relationship curve at the downstream outlet section are used as the boundary conditions of the model, and the lateral inflow or rainfall intensity distribution is set according to the simulation requirements.

[0075] The governing equations are constructed based on the Saint-Venant equations, and the water level and flow velocity of each computational unit are solved simultaneously in each computational time step using the finite difference method.

[0076] The calculation is performed iteratively according to the time step, simulating the spatiotemporal evolution of hydraulic elements in the entire target river channel, and outputting real-time hydrodynamic data including the vector distribution of water flow velocity, water level elevation, dynamic inundation range and water depth of each river section.

[0077] The hydrodynamic data of the entire field is transmitted in real time to the riverbed simulation module 23 and the event generation module 3 through a standardized data interface, serving as their driving data;

[0078] Riverbed simulation module 23 is used to simulate the sediment erosion and deposition process of the riverbed based on the hydrodynamic simulation results of hydrodynamic simulation module 22. It defines the river section parameters of sediment erosion, transport, and deposition flux in each river section during the simulation period, and evaluates the structural stability coefficient of the current riverbank parameters under different hydrological conditions. The evaluation process of the structural stability coefficient is as follows:

[0079] Receive the output results of the full field hydrodynamic data of each river section from the hydrodynamic simulation module 22, as well as the riverbed and embankment status data;

[0080] Based on the limit equilibrium method, the anti-sliding stability analysis of the dike slope is carried out to evaluate its anti-sliding safety factor under adverse conditions such as flood soaking and sudden drop in water level.

[0081] Conduct seepage stability analysis of the embankment and foundation, assess seepage gradient, and prevent piping and soil erosion.

[0082] For revetment and retaining wall structures, assess their resistance to overturning and sliding.

[0083] Based on the scouring effect of water flow on the toe of the dike and the bank slope, and combined with the scouring and deposition depth simulated by riverbed evolution, the stability of the foundation is assessed.

[0084] Event generation module 3 is used to automatically identify and generate several potential risk events corresponding to each river segment based on the simulation results of hydrological and structural safety of each river segment in the current simulation cycle. Risk events include at least dike breach, bank slope instability, excessive scouring or silting of the riverbed, and ecological degradation, and their trigger thresholds and key parameters can be customized by the user.

[0085] The process of risk event generation is as follows:

[0086] Based on the simulation results received by river simulation unit 2, key assessment parameters corresponding to each risk event are extracted, including:

[0087] For the risk of dike breach, key assessment parameters include the difference between the water level and the top elevation of the dike, and the dike stability coefficient.

[0088] For the risk of bank slope instability, key assessment parameters include the slope safety factor and the rate of water level change;

[0089] For the risk of excessive scouring or silting of the riverbed, key assessment parameters include the depth of scouring or the thickness of siltation.

[0090] For the risk of ecological degradation, key assessment parameters include water flow velocity, water depth, and duration of inundation.

[0091] The extracted key evaluation parameters are compared with user-defined trigger thresholds. The trigger thresholds are set separately for each type of risk event, and the trigger conditions for exceeding or falling below the thresholds are defined according to the nature of the parameters.

[0092] When any key assessment parameter meets its triggering condition, the corresponding risk event identifier is automatically generated, and a list of potential risk events for that river segment during the current simulation period is summarized and output.

[0093] The scheme management module 4 is used to acquire and store preset treatment schemes for handling measures and engineering parameters for one or more risk events in the target river section.

[0094] The indicator extraction module 5 is used to quantitatively extract several impact indicators of the governance schemes preset in the scheme management module 4 on various risk events, and simultaneously evaluate the implementation difficulty indicators of the schemes. The impact indicators include engineering cost, expected effect intensity, and ecological disturbance degree, while the implementation difficulty indicators include technical complexity and construction period.

[0095] Intervention prediction module 6 is used to input all the impact indicators and implementation difficulty indicators output by indicator extraction module 5 into the risk intervention model preset by machine learning. The model jointly predicts the intervention coefficient of each indicator on the corresponding risk event in the future period. The intervention coefficient characterizes the ability of the governance plan to suppress the probability and severity of the risk event.

[0096] The judgment module 7 is used to compare the future interference coefficients output by the interference prediction module 6 with the preset standard thresholds. When the interference coefficient of any risk event fails to meet its standard threshold, it is determined that the current governance plan has not met the standard.

[0097] The scheme adjustment module 8 is used to construct a correlation state network between various influencing indicators based on historical data when the scheme fails to meet the standards. The correlation state includes the mutual promotion, inhibition or coupling relationship between the indicators. Based on the correlation state network, with the goal of maximizing the total interference coefficient of the system, the optimal adjustment range of the influencing indicators that fail to meet the standards is solved, and the adjusted governance scheme is generated as an optimization reference.

[0098] The management module 1 is connected to the river simulation unit 2, the event generation module 3, and the scheme management module 4 via a wireless network. The index extraction module 5 is connected to the event generation module 3, the scheme management module 4, and the interference prediction module 6 via a wireless network. The judgment module 7 is connected to the interference prediction module 6 and the scheme adjustment module 8 via a wireless network.

[0099] Compared with existing technologies, the advantage lies in its ability to intuitively preview the long-term effects of a solution under various risk scenarios before implementation, and to automatically identify the shortcomings of the solution. Through intelligent algorithms, it provides accurate optimization suggestions that comprehensively consider safety, economic and ecological benefits, thereby improving the scientific and economic efficiency of the governance solution and effectively avoiding decision-making errors and waste of resources.

[0100] At other levels, the construction process of a risk intervention model in this embodiment is as follows:

[0101] Collect and organize historical governance scheme datasets. Each sample data includes: input feature vector: quantitative values ​​of various impact indicators and implementation difficulty indicators contained in the sample governance scheme; output label vector: after the implementation of the sample governance scheme, the actual intervention coefficient for various risk events obtained through actual monitoring or high-precision post-event simulation. The actual intervention coefficient is comprehensively quantified by the percentage reduction in the probability of the risk event and the percentage reduction in expected loss.

[0102] The input feature vectors are standardized, and feature selection is performed based on feature importance analysis. The random forest algorithm is selected as the base model.

[0103] The constructed dataset is divided into a training set and a test set according to a preset ratio. The selected algorithm is trained using the training set with the goal of minimizing the error between the predicted interference coefficient and the actual interference coefficient. The trained model is validated using the test set, and mean squared error and coefficient of determination are used as evaluation metrics for model performance.

[0104] The validated risk intervention model is deployed in the intervention prediction module 6 to predict the intervention coefficient of the newly input governance scheme. At the same time, the system reserves a data interface to receive the aftereffect data of the newly implemented governance scheme and to start the model retraining process periodically or triggered to realize the dynamic update and adaptive optimization of the model.

[0105] The formula for calculating the interference coefficient of the risk intervention model for any risk event is expressed as:

[0106] ;

[0107] In the formula, Represents risk events The future interference coefficient characterizes the ability of the governance scheme to suppress the probability of the occurrence of this risk event. This represents the logistic function, used to map linear combinations to probability values. Represents risk events The model bias term is learned through training data. This represents the total number of influencing indicators. Representing the Each indicator for risk events The weighting coefficients, Representing the Standardized values ​​for indicators that affect performance or are difficult to implement. Representing the The first indicator and the first The interaction weighting coefficients between the indicators are used to capture the joint effect among them and are learned through training data. Representing the The standardized values ​​of the influencing indicators, Representing the The standardized value of each implementation difficulty indicator.

[0108] This embodiment provides a working logic for the scheme adjustment module 8, specifically as follows:

[0109] Based on the current values ​​of all influencing indicators, an adjustable search space is defined for each indicator, thereby initializing a population containing multiple potential adjustment schemes;

[0110] Entering the iterative optimization loop, in each generation, the adjustment plan represented by each individual in the population is substituted into the risk intervention model, the intervention coefficient of all risk events is re-predicted, and the comprehensive implementation difficulty corresponding to each plan is calculated.

[0111] Based on the overall implementation difficulty, the adaptability of each adjustment plan is evaluated. The core evaluation criterion is to maximize the sum of the interference coefficients of all risk events while minimizing the overall implementation difficulty of the plan, on the premise that the interference coefficients of all risk events reach the standard threshold.

[0112] Individuals in the population are sorted and selected based on the fitness of each adjustment scheme. The preferred schemes with fitness higher than a preset threshold are retained. New possible solutions are explored by performing crossover and mutation operations on the preferred schemes, which are similar to those in biological genetics. In this process, the probability acceptance criterion in the simulated annealing mechanism is introduced to accept some solutions that do not meet the constraints with a certain probability, thereby maintaining the diversity of the population and avoiding the optimization process from getting trapped in local optima too early.

[0113] Repeat the above selection, crossover and mutation steps until the preset number of iterations or fitness convergence criteria are reached.

[0114] All non-dominated solutions that satisfy the constraints are selected from the last generation of the population to form a Pareto optimal solution set. Each solution in this set represents the optimal adjustment range of a set of non-compliant impact indicators, serving as a reference for multiple alternative optimizations of the current governance scheme.

[0115] In summary, this invention constructs a digital twin model to comprehensively simulate the real effects of governance solutions in virtual space, thereby proactively mitigating and significantly reducing decision-making risks. Through quantitative simulation, this invention can accurately predict the complex impacts of different engineering measures on hydrodynamics, riverbed evolution, and ecology, effectively avoiding engineering failures or negative effects caused by design flaws, and fundamentally saving enormous trial-and-error costs.

[0116] Under multiple constraints such as safety, economy, and ecology, the efficient search for Pareto fronts of non-dominated solutions provides decision-makers with a scientific basis for comparing multiple options rather than a single solution, greatly improving the scientific and comprehensive nature of decision-making. Through intuitive three-dimensional dynamic simulation and clear indicator comparison, it effectively promotes communication and consensus among different departments and the public, making the formulation process of governance solutions more transparent, ensuring that the selected solutions achieve the optimal balance in terms of technical, economic, and social benefits, and realizing the sustainable development governance of the watershed.

[0117] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A system for optimizing river risk simulation and management schemes based on scheme pre-simulation, characterized in that, include: The management module is used to obtain access permissions for various river and hydrological data of the target water area, and provides editing permissions for the control command data stream of each functional module; The river channel simulation unit is used to divide the target river channel into several river sections and virtually construct a three-dimensional model based on the actual water channel and hydrological data. It simulates water flow velocity, water level, inundation range and water depth based on hydrological data, and simulates riverbed sediment state and riverbank strength data based on river channel data. The event generation module is used to automatically identify and generate several potential risk events corresponding to each river segment based on the simulation results of the hydrological and structural safety of each river segment in the current simulation cycle. The scheme management module is used to acquire and store preset treatment schemes, including the treatment measures and engineering parameters to be taken for one or more risk events in a target river section. The indicator extraction module is used to quantitatively extract several impact indicators of the governance plans preset by the plan management module on various risk events, and simultaneously evaluate the implementation difficulty indicators of the plans. The intervention prediction module is used to input all the impact indicators and implementation difficulty indicators output by the indicator extraction module into the risk intervention model preset by machine learning. The model jointly predicts the intervention coefficient of each indicator on the corresponding risk event in the future period. The judgment module is used to compare the future interference coefficients output by the interference prediction module with the preset standard thresholds. When the interference coefficient of any risk event fails to meet its standard threshold, the current governance plan is judged to be substandard. The scheme adjustment module is used to construct a correlation state network between various influencing indicators based on historical data when the scheme fails to meet the standards. Based on the correlation state network, with the goal of maximizing the total system interference coefficient, the module solves for the optimal adjustment range of the influencing indicators that fail to meet the standards and generates an adjusted governance scheme as an optimization reference.

2. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 1, characterized in that, The river channel simulation unit has sub-modules deployed at its lower levels. These sub-modules include a river segment construction module, a hydrodynamic simulation module, and a riverbed simulation module. The river segment construction module interacts with the hydrodynamic simulation module and the riverbed simulation module via a wireless network. The river segment construction module is used to divide the river into several continuous river segments based on the actual geographic and hydrological data of the target river, and to build a digital twin model for each river segment that includes three-dimensional topography, riverbed quality, levee structure and hydraulic roughness parameters. The hydrodynamic simulation module is used to import boundary conditions from hydrological data, run the hydrodynamic model on the digital twin model, and simulate the spatial distribution and temporal evolution of water flow velocity, water level, inundation range and water depth in each river section. The riverbed simulation module is used to simulate the sediment erosion and deposition process of the riverbed based on the hydrodynamic simulation results of the hydrodynamic simulation module. It defines the river section parameters of sediment erosion, transport and deposition flux in each river section during the simulation period, and evaluates the structural stability coefficient of the current riverbank parameters under different hydrological conditions.

3. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 2, characterized in that, The actual geographic and hydrological data in the river section construction module include lidar point clouds, UAV aerial survey images, multibeam underwater topographic survey data, historical hydrological observation data, and geological exploration data.

4. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 2, characterized in that, The process of constructing the hydrodynamic model in the hydrodynamic simulation module is as follows: Based on the river section construction module, three-dimensional terrain data and river cross-section data are obtained, and grids are generated to form a grid of two-dimensional computing units. Each computing unit is assigned an initial water level and an initial flow velocity. The flow process curve at the upstream inlet section and the water level-flow relationship curve at the downstream outlet section are used as the boundary conditions of the model, and the lateral inflow or rainfall intensity distribution is set according to the simulation requirements. The governing equations are constructed based on the Saint-Venant equations, and the water level and flow velocity of each computational unit are solved simultaneously in each computational time step using the finite difference method. The calculation is performed iteratively according to the time step, simulating the spatiotemporal evolution of hydraulic elements in the entire target river channel, and outputting real-time hydrodynamic data including the vector distribution of water flow velocity, water level elevation, dynamic inundation range and water depth of each river section. The hydrodynamic data from the entire field is transmitted in real time to the riverbed simulation module and the event generation module through a standardized data interface, serving as their driving data.

5. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 2, characterized in that, The process by which the riverbed simulation module evaluates the structural stability coefficient is as follows: Receive the output results of the full-field hydrodynamic data of each river section from the hydrodynamic simulation module, as well as the riverbed and embankment status data; Based on the limit equilibrium method, the anti-sliding stability analysis of the dike slope is carried out to evaluate its anti-sliding safety factor under adverse conditions such as flood soaking and sudden drop in water level. Conduct seepage stability analysis of the embankment and foundation, and assess the seepage gradient; For revetment and retaining wall structures, assess their resistance to overturning and sliding. Based on the scouring effect of water flow on the toe of the dike and the bank slope, and combined with the scouring and deposition depth simulated by riverbed evolution, the stability of the foundation is assessed.

6. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 1, characterized in that, The process by which the event generation module generates risk events is as follows: Based on the simulation results received by the river simulation unit, key assessment parameters corresponding to each risk event are extracted; The extracted key evaluation parameters are compared with user-defined trigger thresholds. The trigger thresholds are set separately for each type of risk event, and the trigger conditions for exceeding or falling below the thresholds are defined according to the nature of the parameters. When any key assessment parameter meets its triggering condition, the corresponding risk event identifier is automatically generated, and a list of potential risk events for that river segment during the current simulation period is summarized and output.

7. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 1, characterized in that, The construction process of the risk intervention model in the intervention prediction module is as follows: Collect and organize a dataset of historical governance schemes. Each sample data includes: an input feature vector and an output label vector. The input feature vectors are standardized, and feature selection is performed based on feature importance analysis. The random forest algorithm is selected as the base model. The constructed dataset is divided into a training set and a test set according to a preset ratio. The selected algorithm is trained using the training set with the goal of minimizing the error between the predicted interference coefficient and the actual interference coefficient. The trained model is then validated using the test set. The validated risk intervention model is deployed in the intervention prediction module to predict the intervention coefficient for new input governance schemes.

8. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 1, characterized in that, The formula for calculating the interference coefficient of the risk interference model in the interference prediction module for any risk event is expressed as follows: ； In the formula, Represents risk events The future interference coefficient, Represents the logistic function. Represents risk events The model bias term, This represents the total number of influencing indicators. Representing the Each indicator for risk events The weighting coefficients, Representing the Standardized values ​​for indicators that affect performance or are difficult to implement. Representing the The first indicator and the first The interaction weighting coefficients between the indicators Representing the The standardized values ​​of the influencing indicators, Representing the The standardized value of each implementation difficulty indicator.

9. The river risk simulation and management scheme optimization system based on scheme pre-simulation as described in claim 1, characterized in that, The working logic of the scheme adjustment module is as follows: Based on the current values ​​of all influencing indicators, an adjustable search space is defined for each indicator, thereby initializing a population containing multiple potential adjustment schemes; Entering the iterative optimization loop, in each generation, the adjustment plan represented by each individual in the population is substituted into the risk intervention model, the intervention coefficient of all risk events is re-predicted, and the comprehensive implementation difficulty corresponding to each plan is calculated. Based on the overall implementation difficulty, assess the suitability of each adjustment plan; Individuals in the population are sorted and selected based on the fitness of each adjustment scheme, and the preferred schemes with fitness higher than a preset threshold are retained. The preferred schemes are then subjected to crossover and mutation operations that simulate biological genetics. Repeat the above selection, crossover and mutation steps until the preset number of iterations or fitness convergence criteria are reached. All non-dominated solutions that satisfy the constraints are selected from the last generation of the population to form a Pareto optimal solution set, which serves as a reference for multiple alternative optimizations of the current governance scheme.

10. The river risk simulation and management scheme optimization system based on scheme pre-simulation according to claim 1, characterized in that, The management module interacts with the river simulation unit, the event generation module, and the scheme management module via a wireless network. The index extraction module interacts with the event generation module, the scheme management module, and the interference prediction module via a wireless network. The judgment module interacts with the interference prediction module and the scheme adjustment module via a wireless network.

Images