A flood deduction method based on human-computer interaction can be embedded in embankment breach in real time

Abstract

The application discloses a flood deduction method based on human-computer interaction and capable of embedding embankment breach in real time, and the method comprises the following steps: step 1, acquiring basic data; step 2, deducing regional grid discretization; step 3, model parameter assignment and initialization; step 4, model flux calculation; step 5, model thread setting; step 6, model encapsulation; and step 7, terminal human-computer interaction setting. The method can add a new breach in real time in the model calculation process, adopts a double-thread technology to transfer the breach information to a main thread of a running model in real time, realizes uninterrupted and continuous simulation of a water flow process, overcomes the defect that an existing embankment breach flood deduction mode can only fix a breach position in advance, and provides a new and more flexible flood deduction mode for intelligent water conservancy informatization construction.

Description

Technical Field

[0001] This invention relates to the field of flood control and disaster reduction technology in water conservancy projects, specifically to a flood simulation method based on human-computer interaction that can be embedded in real time into dike breaches. Background Technology

[0002] Currently, a new round of industrial revolution characterized by digitalization and intelligence is flourishing. The Ministry of Water Resources attaches great importance to the construction of smart water conservancy, clarifying that the main tasks of smart water conservancy construction are to build digital twin river basins, construct a "2+N" business application system with "four prediction" functions, and improve the network security protection system. The so-called "four predictions" refer to forecasting, early warning, pre-drilling, and contingency planning, and proposes to realize them through digital scenarios, intelligent simulation, and precise decision-making. A smart flood control system that meets the "four prediction" functions has become a significant symbol of high-quality development in water conservancy, and of course, it also places higher demands on the water conservancy business models used in the system.

[0003] Hydrodynamic models are a crucial component of water conservancy operational models and a primary technical tool for flood simulation. Simulating levee breach floods is a common and important scenario in flood simulation. Currently, the mainstream technical tool is a one-dimensional / two-dimensional coupled hydrodynamic model, where the river channel is simulated using a one-dimensional model and the inundation zone using a two-dimensional model. The fixed location and timing of the breach are pre-set, and the one-dimensional and two-dimensional models are coupled through the breach. However, with the increasing demands of flood control technology support operations, the inherent shortcomings of this method have become increasingly apparent. Firstly, this coupled calculation method, due to early limitations in computing resources, adopted a loose coupling between the river channel and the inundation zone to ensure a certain calculation speed, only considering mass exchange between the one-dimensional and two-dimensional models, without considering momentum exchange. Secondly, this method requires fixing the breach location at the beginning, making it impossible to add breaches in real-time according to the simulation requirements. With the continuous development of smart water conservancy information technology, this fixed-breach-location levee breach flood simulation model can no longer fully meet current operational requirements. Therefore, it is both necessary and urgent to study a flood simulation method based on human-computer interaction that can be embedded in real time to simulate dike breaches. Summary of the Invention

[0004] The purpose of this invention is to propose a flood prediction method based on human-computer interaction that can be embedded in real time into dike breaches, in response to the current development needs of smart water conservancy information technology.

[0005] This invention is achieved through the following technical solution:

[0006] This invention discloses a flood simulation method based on human-computer interaction and real-time embedding of dike breach simulation. The method employs a two-dimensional unstructured hydrodynamic model to simulate flood movement across the entire flood simulation area, which includes the river channel and the inundated zone. After the model starts, a dual-threaded technique is used, allowing breach conditions to be set at any dike location on the terminal interface during the calculation process. The model automatically loads new breach conditions and continues the simulation of the subsequent continuous two-dimensional flood movement process under the current flood movement situation. Specifically, the method includes the following steps:

[0007] Step 1: Acquisition of basic data: Acquire basic data, including topographic data, land use type data, spatial location and top elevation data of dikes in the simulation area, upstream inflow process data used in flood simulation, and downstream water level-discharge relationship data.

[0008] Step 2: Discretization of the simulation area grid: The entire flood simulation area is discretized uniformly using a triangular unstructured grid. The dike line is used as the control line for grid partitioning. After the control line is discretized, it is used as the edge of the triangular grid cell. This type of cell edge is called the dike edge, and other edges are called ordinary edges. The effect of breaching is simulated by the change of the dike edge elevation.

[0009] Step 3, Model Parameter Assignment and Initialization: First, the elevation of each grid cell is assigned using topographic data, and the elevation value is stored at the center of the grid cell; the elevation of the dike edge is assigned using the dike crest elevation data, and the initial elevation of the dike edge is the dike crest elevation data by default, while the elevation of other ordinary edges is uniformly assigned to zero; the roughness parameter of each grid cell is assigned according to the land use type data; the initial values ​​of water depth and flow velocity of each grid cell are assigned, the initial values ​​of grid cells in the river channel are calculated based on the initial values ​​of the river inflow process data, and the initial values ​​of grid cells in the flooded area are uniformly assigned to zero;

[0010] Step 4: Model Flux Calculation: A two-dimensional hydrodynamic model is constructed using the finite volume method based on the Godunov scheme. Normal flux calculations are performed on ordinary edges using the Roe scheme, while the normal numerical flux of the levee edges is calculated using the following formula:

[0011] (2)

[0012] In the formula: It is the vector form of the normal numerical flux of the dike side; The unit width flow rate along the dike; The water depth at the edge of the dike; g is the acceleration due to gravity;

[0013] Step 5, Model Thread Setup: The model is developed using dual-thread technology. Thread 1 is the main thread, which performs the two-dimensional hydrodynamic model calculation normally. Thread 2 performs input detection to check whether there is a change in the elevation of the embankment. During the execution of Thread 1, Thread 2 continuously detects whether there is user command input.

[0014] Step 6, Model Packaging: Set the program to run under the Windows operating system and package the two-dimensional hydrodynamic model into a console executable program;

[0015] Step 7, Terminal Human-Computer Interaction Settings: During the calculation process, click on any dike side on the terminal interface to set the bottom elevation data of the breach after the selected dike side collapses. After clicking to confirm, a new dike side elevation file is generated, including the dike side number and the new dike side elevation information. At the same time, a command to input a change in dike side elevation is triggered to the executable program on the console. After thread 2 detects the command, it notifies thread 1. Thread 1 reads the newly generated dike side elevation file in real time, completing the non-blocking input of breach information during the model calculation process, thereby realizing the simulation of the effect of adding a breach.

[0016] Repeat step 7 to achieve the simulation effect of arbitrarily embedding multiple breaches in a flood simulation process until the calculation ends.

[0017] Furthermore, the terrain data mentioned in step 1 is obtained through surveying and mapping methods. Specifically, a UAV equipped with a LiDAR sensor is used to obtain ground elevation data above the water surface in the simulation area, and a survey vessel equipped with an ADCP sensor is used to obtain terrain elevation data below the water surface. The two parts of terrain data are fused to obtain the overall terrain data of the simulation area; the resolution of the terrain data is not less than 10m.

[0018] The acquisition of land use type data specifically involves obtaining free 2.5m resolution remote sensing image data via the internet, and then interpreting it to obtain land use type data for the simulation area.

[0019] The spatial location and top elevation data of the dike were obtained through dike design drawings or on-site surveying.

[0020] The flood simulation uses upstream inflow data and downstream water level-discharge relationship data obtained from the hydrological department's real-time hydrological and rainfall database.

[0021] Furthermore, the governing equations used in the two-dimensional hydrodynamic model in step 4 are as follows:

[0022] (1)

[0023] In the formula: ;

[0024] h represents the water depth; u and v represent the flow velocities in the x and y directions, respectively; and t represents time. The slopes are in the x and y directions, respectively; Z b Where is the ground elevation; g is the acceleration due to gravity; Let n be the friction term in the x and y directions, respectively, where n is the Manning roughness coefficient.

[0025] Furthermore, in formula (2) of step 4, the unit width flow rate q of the dike side and the water depth h on the dike side are... * The solution formula is:

[0026] (3)

[0027] (4)

[0028] In the formula: Z up Z down These represent the water levels of adjacent upstream and downstream units along the dike, with the side with the higher water level being the upstream side; Z levee This refers to the elevation of the dike edge.

[0029] Furthermore, the non-blocking input mentioned in step 7 means that thread 2 periodically checks if there is any input available; if there is, it processes it; if not, it continues to the next round of checks without stopping to wait or pausing or stopping the normal execution of thread 1.

[0030] The beneficial effects of this invention are: the method described in this invention can add new breaches in real time during the model calculation process, and use dual-thread technology to transmit the breach information to the running model main thread in real time, realizing uninterrupted continuous simulation of the water flow process. This overcomes the defect of existing flood prediction models that can only fix the breach location in advance, and provides a new and more flexible flood prediction model for the construction of smart water conservancy information.

[0031] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0032] Figure 1 This is a flowchart of the method described in this invention;

[0033] Figure 2 This is a schematic diagram of the two-dimensional unstructured mesh discretization of the derivation region in this invention;

[0034] Figure 3 This is a calculation effect diagram of real-time embedded rupture in Embodiment 1 of the present invention. Detailed Implementation

[0035] This invention discloses a flood simulation method based on human-computer interaction that allows for real-time embedding of dike breaches. The method discretizes the entire flood simulation area (including the river channel and the inundated area) using a unified two-dimensional unstructured model. The river channel and the inundated area are simulated using a unified two-dimensional hydrodynamic model. The effect of adding breaches is uniformly achieved by modifying the elevation of the dike edges. To add new breaches in real-time during the calculation process, a dual-thread technique is employed. Thread one performs normal calculations, while thread two continuously detects whether new breach information is transmitted to the model. Once new breach input information is detected, thread one automatically reads this information and continues the subsequent coherent two-dimensional flood movement process simulation under the current flood movement situation. This achieves the flood simulation effect of embedding breaches in real-time at any dike location during flood simulation. Figure 1 As shown, the method includes the following steps:

[0036] Step 1: Acquiring basic data:

[0037] Acquiring basic data mainly includes topographic data, land use type data, spatial location and top elevation data of dikes within the simulation area, upstream inflow process data used in flood simulation, and downstream water level-discharge relationship data.

[0038] Specifically, topographic data of the projection area is obtained through surveying and mapping methods. Specifically, a UAV equipped with a LiDAR sensor acquires ground elevation data above the water surface, while a survey vessel equipped with an ADCP sensor acquires subsurface elevation data. These two sets of data are then fused to obtain the overall DEM topographic data for the projection area, with a resolution of at least 10m. Free 2.5m resolution remote sensing imagery is obtained online, and after interpretation, land use type data is obtained. This land use type data is primarily used to assign roughness parameters to the discretized grid cells. Spatial location data and crest elevation data of the dikes are obtained through dike design drawings or on-site surveying. The upstream inflow process data and downstream water level-discharge relationship data used in flood projection can generally be obtained from the real-time hydrological and rainfall database of the hydrological department.

[0039] Step 2: Deducing the region grid discretization:

[0040] A triangular unstructured mesh is used to uniformly discretize the entire flood projection area, balancing computational accuracy and efficiency. The spatial scale of the mesh is 10-50m. During meshing, the river channel and embankment are generalized as a line, which serves as the control line for meshing. After discretization, the control line becomes the edge of the triangular mesh cell; this type of edge is called the embankment edge, while other mesh cell edges are called ordinary edges. Specifically, as follows... Figure 2 As shown. In subsequent calculations, the effect of a breach was simulated by adjusting the elevation of the dike side.

[0041] Step 3: Assigning and initializing model parameters:

[0042] First, the elevation of each grid cell is assigned using topographic data within the simulation area, and the elevation value is stored at the center of the grid cell. The elevation of the dike edges is assigned using the dike crest elevation data; the initial elevation of the dike edges is defaulted to the dike crest elevation data, while the elevations of other ordinary edges are uniformly assigned to zero. The roughness parameter of each grid cell is assigned based on land use type data. Initial values ​​for water depth and flow velocity are assigned to each grid cell within the simulation area. The initial values ​​for grid cells within the river channel can be calculated based on the initial values ​​of the river inflow process data, while the initial values ​​for water depth and flow velocity for grid cells in the flooded area are generally uniformly assigned to zero.

[0043] Step 4: Model flux calculation:

[0044] The river channel and inundation zone within the simulation area are uniformly simulated using a two-dimensional hydrodynamic model. The governing equations used in the two-dimensional hydrodynamic model are as follows:

[0045] (1)

[0046] In the formula: ;

[0047] h represents the water depth; u and v represent the flow velocities in the x and y directions, respectively; and t represents time. The slopes are in the x and y directions, respectively; Z b Where is the ground elevation; g is the acceleration due to gravity; Let n be the friction term in the x and y directions, respectively, where n is the Manning roughness coefficient.

[0048] A two-dimensional hydrodynamic model is constructed using the finite volume method based on the Godunov scheme. The Godunov scheme, which is based on solving the Riemann approximate solution, is the mainstream computational scheme in the field of shallow water numerical simulation. Representative schemes include the Roe scheme and the HLLC scheme. Among them, the Roe scheme is used for ordinary edge flux calculation. The specific numerical method of the Roe scheme can be found in the following literature (Zhang Dawei, Numerical Simulation of Dam Breach Flow Based on Godunov Scheme [M]. China Water Resources and Hydropower Press, Beijing, 2014, 12).

[0049] The normal numerical flux along the levee side is calculated using the following formula:

[0050] (2)

[0051] In the formula: It is the vector form of the normal numerical flux of the dike side; The unit width flow rate along the dike; denoted as ρ, where ρ is the water depth at the edge of the dike; g is the acceleration due to gravity.

[0052] In formula (2), the unit width discharge q of the dike side and the water depth h on the dike side are... * The solution formula is:

[0053] (3)

[0054] (4)

[0055] In the formula: Z up Z down These represent the water levels of adjacent upstream and downstream units along the dike, with the side with the higher water level being the upstream side; Z levee This refers to the elevation of the dike edge.

[0056] Step 5, Model Thread Settings:

[0057] The model is developed using a dual-threaded approach. Thread 1 is the main thread, which performs the two-dimensional hydrodynamic model calculations normally. Thread 2 performs input detection to check whether there are changes in the elevation of the levee. While Thread 1 is executing, Thread 2 continuously checks for user command input. During this process, the execution of Thread 2 does not block the execution of Thread 1.

[0058] Step 6, Model Encapsulation:

[0059] When integrating a model with an information system, the model is typically packaged into a standalone executable file, and then interacts with the system through input and output files. Without loss of generality, this example assumes the program runs on a Windows operating system and packages the two-dimensional hydrodynamic model into a console executable program.

[0060] Step 7, Terminal Human-Computer Interaction Settings:

[0061] During the calculation, clicking on any dike side in the terminal interface sets the bottom elevation data of the breach after the selected dike side collapses. After confirming the click, a new text file containing the dike side elevation, including the dike side number and the new elevation information, is generated. Simultaneously, a command to change the dike side elevation is triggered and input to the executable program in the console. Thread 2 detects this command and notifies Thread 1. Thread 1 then reads the newly generated dike side elevation file in real time, completing the non-blocking input of breach information during the model calculation process, thus simulating the effect of a newly added breach. Non-blocking input means that Thread 2 periodically checks if any input is available. If so, it processes it; otherwise, it continues to the next round of checks without stopping or waiting, and without pausing or stopping the normal execution of Thread 1.

[0062] By repeating step 7, multiple breaches can be arbitrarily embedded in the simulation of a flood until the calculation is completed.

[0063] Example 1

[0064] This embodiment is a specific application example of the above method.

[0065] The computational domain in this embodiment is a physical model of the Toce River, approximately 5 km upstream, built by ENEL in Milan, Italy, at a scale of 1:100, with a flooded area in the middle of the river channel. A detailed description of this model can be found in the following literature (Zhang Dawei, Research on Mathematical Models and Applications of Flow in Dam Breach [D], Tsinghua University Doctoral Dissertation, 2008). Unstructured discretization was performed using 28,146 triangular mesh elements, with a uniform roughness value of 0.0162. A two-dimensional hydrodynamic model was developed using C++, and a test interface was developed using Python.

[0066] The calculation effect diagram of the real-time embedded vulnerability in this embodiment is shown below. Figure 3 As shown. When the calculation progress reaches 72%, a breach is added through the interface. Specifically, the elevation of the three dike sides numbered 397, 713, and 877 is uniformly changed to 7.31m. Click... Figure 3 After clicking the "Sent" button on the left, water flowed from the river channel into the flooded area through the newly added breach, successfully adding the breach effect. Figure 3 The upper right half shows a magnified view of the water flowing into the flooded area. Figure 3 The lower right half shows the complete simulation area, demonstrating that the boundary processing method proposed in this invention has universality.

[0067] The above embodiments are only a partial embodiment of the present invention and do not cover all of the present invention. Based on the above embodiments and the accompanying drawings, those skilled in the art can obtain more implementation methods without creative effort. Therefore, all implementation methods obtained without creative effort should be included within the protection scope of the present invention.

Claims

1. A flood prediction method based on human-computer interaction that can be embedded in real time into dike breaches, characterized in that: The method employs a two-dimensional unstructured hydrodynamic model to simulate flood movement across the entire flood simulation area, which includes the river channel and the inundated zone. After the model starts, a dual-threaded technique is used, allowing breach conditions to be set at any levee location on the terminal interface during the calculation process. The model automatically loads new breach conditions and continues to simulate the subsequent continuous two-dimensional flood movement process under the current flood movement situation. Specifically, the method includes the following steps: Step 1: Acquisition of basic data: Acquire basic data, including topographic data, land use type data, spatial location and top elevation data of dikes in the simulation area, upstream inflow process data used in flood simulation, and downstream water level-discharge relationship data. Step 2: Discretization of the simulation area grid: The entire flood simulation area is discretized uniformly using a triangular unstructured grid. The dike line is used as the control line for grid partitioning. After the control line is discretized, it is used as the edge of the triangular grid cell. This type of cell edge is called the dike edge, and other edges are called ordinary edges. The effect of breaching is simulated by the change of the dike edge elevation. Step 3, Model Parameter Assignment and Initialization: First, the elevation of each grid cell is assigned using topographic data, and the elevation value is stored at the center of the grid cell; the elevation of the dike edge is assigned using the dike crest elevation data, and the initial elevation of the dike edge is the dike crest elevation data by default, while the elevation of other ordinary edges is uniformly assigned to zero; the roughness parameter of each grid cell is assigned according to the land use type data; the initial values ​​of water depth and flow velocity of each grid cell are assigned, the initial values ​​of grid cells in the river channel are calculated based on the initial values ​​of the river inflow process data, and the initial values ​​of grid cells in the flooded area are uniformly assigned to zero; Step 4: Model Flux Calculation: A two-dimensional hydrodynamic model is constructed using the finite volume method based on the Godunov scheme. Normal flux calculations are performed on ordinary edges using the Roe scheme, while the normal numerical flux of the levee edges is calculated using the following formula: （2） In the formula: It is the vector form of the normal numerical flux of the dike side; The unit width flow rate along the dike; The water depth at the edge of the dike; g is the acceleration due to gravity; Step 5, Model Thread Setup: The model is developed using dual-thread technology. Thread 1 is the main thread, which performs the two-dimensional hydrodynamic model calculation normally. Thread 2 performs input detection to check whether there is a change in the elevation of the embankment. During the execution of Thread 1, Thread 2 continuously detects whether there is user command input. Step 6, Model Packaging: Set the program to run under the Windows operating system and package the two-dimensional hydrodynamic model into a console executable program; Step 7, Terminal Human-Computer Interaction Settings: During the calculation process, click on any dike side on the terminal interface to set the bottom elevation data of the breach after the selected dike side collapses. After clicking to confirm, a new dike side elevation file is generated, including the dike side number and the new dike side elevation information. At the same time, a command to input a change in dike side elevation is triggered to the executable program on the console. After thread 2 detects the command, it notifies thread 1. Thread 1 reads the newly generated dike side elevation file in real time, completing the non-blocking input of breach information during the model calculation process, thereby realizing the simulation of the effect of adding a breach. Repeat step 7 to achieve the simulation effect of arbitrarily embedding multiple breaches in a flood simulation process until the calculation ends.

2. The flood projection method based on human-computer interaction and capable of real-time embedding into dike breach simulation as described in claim 1, characterized in that, The terrain data mentioned in step 1 is obtained through surveying and mapping methods. Specifically, a UAV equipped with a LiDAR sensor is used to obtain ground elevation data above the water surface in the simulation area, and a survey vessel equipped with an ADCP sensor is used to obtain terrain elevation data below the water surface. The two parts of terrain data are fused to obtain the overall terrain data of the simulation area. The resolution of the terrain data is not less than 10m. The acquisition of land use type data specifically involves obtaining free 2.5m resolution remote sensing image data via the internet, and then interpreting it to obtain land use type data for the simulation area. The spatial location and top elevation data of the dike were obtained through dike design drawings or on-site surveying. The flood simulation uses upstream inflow data and downstream water level-discharge relationship data obtained from the hydrological department's real-time hydrological and rainfall database.

3. The flood projection method based on human-computer interaction and capable of real-time embedding into dike breach simulation as described in claim 1, characterized in that, The governing equations used in the two-dimensional hydrodynamic model in step 4 are as follows: （1） In the formula: ; h represents the water depth; u and v represent the flow velocities in the x and y directions, respectively; and t represents time. The slopes are in the x and y directions, respectively; Z b Where is the ground elevation; g is the acceleration due to gravity; Let n be the friction term in the x and y directions, respectively, where n is the Manning roughness coefficient.

4. The flood projection method based on human-computer interaction and capable of real-time embedding into dike breach simulation as described in claim 1, characterized in that, In formula (2) of step 4, the unit width flow rate q of the dike side and the water depth h on the dike side are... * The solution formula is: （3） （4） In the formula: Z up Z down These represent the water levels of adjacent upstream and downstream units along the dike, with the side with the higher water level being the upstream side; Z levee This refers to the elevation of the dike edge.

5. The flood simulation method based on human-computer interaction and capable of real-time embedding into dike breach simulation according to claim 1, characterized in that, The non-blocking input mentioned in step 7 means that thread 2 periodically checks if there is any input available; if there is, it processes it; if not, it continues to the next round of checks without stopping to wait or pausing or stopping the normal execution of thread 1.

Images