Efficient lightweight urban inland inundation instant simulation model construction method and system

Abstract

The invention relates to an efficient lightweight urban inland inundation instant simulation model construction method, equipment and a medium, and belongs to the technical field of water environment engineering. In order to solve the problems of pipe network and surface water flow separation, infiltration process simplification, low calculation efficiency and visualization lag in the prior art, the method comprises the following steps: collecting modeling area data and creating terrain, rainfall level increment and infiltration rate texture; constructing a physical parameter enhanced height difference-flow conservation surface water power model PEHDCM; constructing a simplified SWMM underground pipe network model and associating the model with a surface grid to realize bidirectional dynamic coupling of the model and the surface grid; and a real-time multi-channel body rendering pipeline is constructed, and real-time simulation rendering of the earth surface runoff production fluid is realized. According to the invention, efficient simulation and instant visualization of the whole process of urban waterlogging from calculation to rendering are realized.

Description

TECHNICAL FIELD

[0001] The application relates to a high-efficiency lightweight urban waterlogging real-time simulation model construction method and system, and belongs to the technical field of water environment engineering. BACKGROUND

[0002] With the acceleration of urbanization, the problem of urban waterlogging caused by extreme rainfall is increasingly prominent, which not only affects the life and property safety of citizens, but also causes serious threat to urban infrastructure. Accurate and rapid flood simulation and intuitive visualization are the key to flood control decision-making. In the prior art, most models only calculate pipe network drainage or surface water accumulation alone, ignoring the dynamic interaction between the two, and do not consider the dynamic change of soil permeability with water content. A fixed infiltration rate is used, which cannot reflect the spatio-temporal heterogeneity of the actual infiltration process. The establishment of a simplified or complete shallow water equation+SWMM coupled model often requires a large amount of time and labor cost, and has the problems of complex model building, tedious data processing and low calculation efficiency. And the known pipe network hydraulic simulation model, surface flood simulation and model coupled with the two to different degrees, will have the problem of delayed display of simulation results after waiting for the completion of simulation calculation. The simulation display is not timely and intuitive, and cannot reflect the dynamic what you see is what you get. SUMMARY

[0003] The application aims to solve the problems of pipe network and surface water flow separation, simplified infiltration process, low calculation efficiency and delayed result display in the existing urban waterlogging simulation technology, and proposes a high-efficiency lightweight urban waterlogging real-time simulation model construction method and system.

[0004] The technical scheme of the application is as follows:

[0005] The application discloses a high-efficiency lightweight urban waterlogging real-time simulation model construction method, which comprises the following steps: S1, collecting data of a modeling area, including rainfall time series data, digital surface model (DSM) raster data of a lower surface, soil distribution raster data and underground drainage pipe network data; S2, creating terrain height map texture based on the DSM raster data in the step S1; converting and calculating a real water level increment on the ground based on the rainfall time series data in the step S1 to create rainfall-based water level increment texture data; creating infiltration rate texture data based on the soil distribution raster data in the step S1, and generating actual ground water production based on the rainfall-based water level increment and the infiltration rate; S3, constructing a physical parameter enhanced height difference-flow conservation ground water dynamic model (PEHDCM); S4, reconstructing a SWMM pipe network model, retaining only conveying components, constructing the SWMM pipe network model based on the underground drainage pipe network data in the step S1, and creating a grid association mapping table of pipe network model nodes and the terrain height map texture based on the terrain height map texture in the step S2, which is used for recording and identifying a grid where a pipe network node is located; S5, taking the actual ground water production in the step S2 as inflow data of the SWMM pipe network model node, and taking outflow of the SWMM pipe network model node as a water level increment source item in a ground fluid simulation process, and completing bidirectional data interaction every Δt seconds, wherein Δt is a simulation time step of the SWMM model, bidirectional dynamic coupling of the pipe network and ground water flow is realized, and real-time calculation of a ground water level and a flow direction is realized.

[0006] Specifically, in the step S1,

[0007] The spatial resolution of the digital surface model (DSM) raster data of the lower surface is 2 meters;

[0008] The soil distribution raster data is obtained from a soil database or through remote sensing image inversion interpretation, and the spatial resolution is 2 meters; the underground rainwater pipe network data comprises attribute data of node elements and connecting elements; the unit of the rainfall time series data is millimeter / hour, and the rainfall time series data is collected or generated by using a Chicago rain type storm intensity formula.

[0009] Specifically, the specific implementation of the step S2 comprises:

[0010] S2.1. creating terrain height map texture data according to the digital surface model (DSM) raster data of the lower surface, and the pixel size is consistent with the spatial resolution of the DSM, that is, 2*2 meters;

[0011] S2.2. calculating a dynamic infiltration rate of each grid according to the soil distribution raster data, obtaining infiltration rate texture data, and the calculation expression is as follows:

[0012]

[0013] wherein,​ is the permeability at the current water content, is the saturated permeability of the soil, is the current water content, is the residual water content, is the saturated water content, is the soil porosity distribution index;

[0014] S2.3. According to the rainfall time series data, the conversion to the real water level increment data of the modeling area is calculated to obtain the rainfall-based water level increment texture data, and the calculation expression is:

[0015]

[0016] wherein, is the water depth increment, is the rainfall intensity, is the surface water dynamic simulation time step.

[0017] Specifically, the implementation method of step S3 includes the following steps:

[0018] S3.1. According to the terrain height map texture, the surface area of the modeling area is discretized by using a regular grid with the same resolution, and the grid cell state variables include water level and four-direction outflow;

[0019] S3.2. Discrete calculation is performed based on the physical parameter enhanced height difference-flow conservation model PEHDCM, including flow velocity derivation, flow correction, viscosity smoothing, water level update and physical constraint steps;

[0020] S3.2.1. Flow velocity derivation is performed based on water level gradient and gravity;

[0021] S3.2.2. Viscous smoothing processing of water body is achieved by weighted average of adjacent grid flow velocities;

[0022] S3.2.3. Flow correction is performed by fusing height difference, flow velocity and gravity;

[0023] S3.2.4. Water level update calculation is performed with flow velocity and viscosity correction;

[0024] S3.2.5. Physical constraints are applied, including water conservation constraint and flow velocity constraint.

[0025] Specifically, the implementation method of step S4 includes the following steps:

[0026] S4.1. The SWMM pipe network model is reconstructed, and only the conveying components are retained to construct the SWMM rainwater pipe network hydraulic model;

[0027] S4.2. Superimpose the SWMM pipe network model topology data on each grid real space coordinate of the terrain height map texture data to create a grid association mapping table of the pipe network model nodes and the terrain height map texture.

[0028] Specifically, the specific implementation of the bidirectional dynamic coupling in step S5 includes: S5.1. Forward coupling: superimpose the pipe network node overflow flow calculated by the SWMM rainwater pipe network hydraulic model as a source term of the ground surface fluid simulation to the corresponding ground surface grid; S5.2. Reverse coupling: when the ground surface water level is higher than the pipe network node water level, calculate the inflow of the ground surface water flow into the pipe network; S5.3. Data interaction adopts Socket communication, and the format is JSON, which includes a timestamp, a grid ID / node ID, and a flow / level value.

[0029] Specifically, the construction of the multi-channel volume rendering pipeline in step S6 includes: S6.1. Construct a time synchronization module for synchronizing the simulation start time and the calculation start time of each step; S6.2. Construct a terrain and water level initialization buffer module to integrate various data sources to generate an initial water level texture; S6.3. Construct an outflow calculation buffer module based on the initial water level texture; S6.4. Construct a water level correction buffer module combined with the inflow and the inflow correction water level texture of the SWMM rainwater pipe network hydraulic model; S6.5. Construct an outflow optimization buffer module based on the corrected water level; S6.6. Construct a volume rendering module to convert the normalized water level to a real value and render fluid effects; S6.7. Construct a data interaction interface module with the SWMM rainwater pipe network hydraulic model to realize data transmission between the ground and the pipe network.

[0030] Specifically, the normalization processing in the multi-channel volume rendering pipeline is to convert physical quantities into a unified numerical range, and the conversion expression is:

[0031]

[0032] wherein, is the normalized water level; is the real ground water level; and are the minimum and maximum values of the real terrain elevation span of the modeling area.

[0033] Specifically, the texture used in the multi-channel volume rendering pipeline is a two-dimensional pixel array carrying specific data, serving as a data container for GPU calculation.

[0034] A system of a high-efficiency lightweight urban waterlogging real-time simulation model construction method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, and the computer program realizes the steps of the high-efficiency lightweight urban waterlogging real-time simulation model construction method of any one of the above when running.

[0035] The present application has the following beneficial effects:

[0036] By establishing a two-way coupling mechanism of SWMM and a physical parameter enhanced "height difference-flow conservation" model, surface runoff and infiltration, pipe network inflow or overflow are quickly and dynamically simulated and calculated, and real-time calculation, immediate dynamic rendering simulation evolution of waterlogging accumulation are realized through multi-channel volume rendering.The present application effectively reduces the modeling complexity of two-way coupling of surface and underground pipe network, improves the modeling and simulation calculation efficiency and rich visualization dimension, is more immediate and intuitive, reduces the workload and saves the cost. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 The flowchart of the present application is shown in the figure;

[0038] Figure 2 The volume rendering display effect diagram of the embodiment of the present application is shown in the figure;

[0039] Figure 3 The data flowchart of real-time calculation and immediate display of the present application is shown in the figure. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application is further described in detail below in combination with the figures and specific embodiments.

[0041] Embodiment 1:

[0042] As shown in the figure, a high-efficiency lightweight urban waterlogging real-time simulation model construction method comprises the following steps: Figure 1

[0043] S1. Collecting data of the modeling area, including rainfall time series data, digital surface model DSM (with building height) raster data of the underlying surface, soil type distribution data and underground drainage pipe network data;

[0044] S1.1. Collecting raster data of the digital surface model DSM of the modeling area, setting the spatial resolution to 2 meters; obtaining soil distribution raster data of the modeling area from the China soil database, or creating soil distribution raster data by soil spectral characteristics inversion of remote sensing images, setting the same spatial resolution of 2 meters;

[0045] S1.2. Collect the underground rainwater pipe network data in the modeling area, including nodes (inspection wells, rainwater inlets, storage tanks), connections (pipes, water pumps), and attribute data including spatial XY coordinates, elevation, well depth of node elements, upstream and downstream elevations, cross-section size, material, and water pump characteristic curve data of connection elements;

[0046] S1.3. Collect the rainfall time series data in the modeling area, or use the Chicago rain type storm intensity formula to calculate and generate, with units of mm / h (millimeters / hour);

[0047] S2. Create a terrain height map texture based on the DSM (with building height) raster data in step S1; calculate the real water level increment data on the ground surface based on the rainfall time series conversion, and create rainfall-based water level increment texture data; create infiltration rate texture data based on the underlying soil distribution data in step S1. The actual surface water yield is the rainfall minus the infiltration;

[0048] S2.1. Create terrain height map texture data based on the DSM (with building height) raster data in step S1.1, with a pixel size of 2x2 meters, which is the same as the spatial resolution of the DSM;

[0049] S2.2. Calculate the dynamic permeability of each grid (i.e., raster) based on the soil distribution raster data in step S1.1 , to obtain the infiltration rate texture data, with the calculation expression being:

[0050]

[0051] wherein, is the permeability at the current water content, is the saturated permeability of the soil (empirical value, sandy soil , loam , clay ), is the current water content, is the residual water content, is the saturated water content, is the soil porosity distribution index (empirical value, sandy soil , , ; clay , , );

[0052] S2.3. Calculate the real water depth increment data in the modeling area based on the rainfall time series data in S1.3, to obtain rainfall-based water depth increment texture data, with the calculation expression being:

[0053]

[0054] wherein, is the water depth increment, is the rainfall intensity, is the surface water dynamic simulation time step;

[0055] S3. Constructing a physical parameter enhanced "height difference-flow conservation" surface water dynamic model PEHDCM;

[0056] S3.1. Discretizing the surface area of the modeling area with a regular grid of the same resolution according to the terrain height map texture in step S2.1, the grid cell state variables including: water level (current frame grid water depth), four-direction outflow (right outflow), (downward outflow), (left outflow), (upward outflow);

[0057] S3.2. Based on the physical parameter enhanced height difference-flow conservation model PEHDCM discrete calculation, including four core steps of flow velocity derivation, flow correction, water level update, viscous smoothing, and physical constraints;

[0058] S3.2.1. Flow velocity derivation based on water level gradient and gravity, the calculation formula is:

[0059] ,

[0060] wherein, and the first row of grid , flow velocity in the current calculation step in the direction (lateral and longitudinal), is the gravitational acceleration, and are the water levels of the right and lower adjacent grids in the current calculation step, is the time of the current calculation step, is the grid size (resolution), is the simulation time step;

[0061] S3.2.2. Water viscous smoothing processing calculation, viscous damping is achieved by weighted average of adjacent grid flow velocities to avoid flow velocity sudden change, the calculation formula is:

[0062] ,

[0063] wherein, and For , direction (transverse and longitudinal) viscosity smooth flow rate, and is the current grid (the first row and the first column) in the current calculation step transverse and longitudinal flow rate, , , , is the current calculation step of the transverse flow rate of the left, right, lower and right adjacent grids, , , , is the current calculation step of the longitudinal flow rate of the left, right, lower and right adjacent grids, is the viscosity coefficient of water movement;

[0064] S3.2.3. The height difference, flow rate and gravity are fused to correct the flow, and the calculation expression is:

[0065]

[0066] wherein, is the first time step, and the ground grid (the first row and the first column) flows to direction (right, down, left, up); and is the topographic elevation of the ground grid (the first row and the first column) and the adjacent direction grid; is the water level of the adjacent direction grid; (gravity weight coefficient); (flow rate weight coefficient); is the first time step, direction smooth flow rate (take to the right, to the down, to the left, and to the up); is the old flow smooth weight;

[0067] S3.2.4. Water level update calculation with flow rate and viscosity correction, and the calculation formula is:

[0068]

[0069] The last two terms in the formula are corrections for the water level based on the viscosity term, obtained by... , directional velocity gradients are used to simulate water level changes caused by viscosity, ensuring water conservation and physical consistency.

[0070] S3.2.5. Physical constraints, dual limitation of water volume and flow velocity:

[0071] (1) Water conservation constraint:

[0072]

[0073] in, For the first Time step, ground grid (number) Line 1 (column) towards Actual outflow in direction (right, down, left, up); For the first Time step, ground grid (number) Line 1 The absolute water level of the column; For ground grid (the first) Line 1 The actual terrain elevation (in the column); This represents the actual area of ​​the ground grid.

[0074] (2) Flow velocity constraint: If or (If the velocity exceeds the gravitational flow limit), then the velocity is cut off to... To avoid non-physical flow velocities. Among them, and The first Time step, smoothed flow velocity in both horizontal and vertical directions;

[0075] S4. Reconstruct the SWMM network model, retaining only the transport components. Construct the SWMM stormwater network hydraulic model using the underground stormwater network data from step S1. Based on the topographic height map texture from S2, create a grid association mapping table between the network model nodes and the topographic height map texture to record and identify the grid where the network nodes are located.

[0076] S4.1. Reconstruct and modify the publicly available SWMM model, removing the atmospheric component, surface component, and underground component modules from the model, retaining only the transport component, and constructing an SWMM stormwater pipe network hydraulic model;

[0077] S4.2. Based on the real spatial coordinates of each grid in the terrain height map texture data in S2.1, overlay the SWMM network model topology data created in step S4.1 under a unified coordinate system, and create a grid association mapping table between network model nodes and terrain height map textures to identify the grid where the network node is located. Each SWMM node is associated with one grid.

[0078] S5. Use the surface water produced in step S2 as the inflow data of the SWMM network model node in S3, and use the outflow of the SWMM network model node as the water level increment source term in the fluid simulation process. Complete one bidirectional data interaction every Δt (SWMM simulation time step) seconds to realize the bidirectional dynamic coupling between the network and the surface water flow, and calculate the surface water level and flow direction in real time.

[0079] S5.1. Forward Coupling (SWMM to Surface Grid): Overflow flow rate of pipe network nodes calculated by SWMM. (unit: As the source term for surface fluid simulation, it is superimposed onto the corresponding surface grid, and the expression for calculating the water level increment is:

[0080] , ,

[0081] in, For the first Time step, surface grid (the first) Line 1 (List) Water level increase caused by overflow at SWMM network nodes; For the first Time step, SWMM network ordinary node overflow (the first) Line 1 Each grid can contain 0 or more pipe network nodes. The smaller the grid size, the fewer pipe network nodes it can contain, and the more accurate the simulation. For the first Time step, SWMM pump node The drainage volume (as a negative source item, available when the pump node is activated). For time step; For the surface grid (the first Line 1 The actual area of ​​the column; For the first Time step, surface grid (the first) Line 1 The current absolute water level of (the column); For the first Time step, surface grid (the first) Line 1 The new absolute water level (in a grid);

[0082] S5.2. Reverse Coupling (Surface Mesh to SWMM): If the surface water level obtained from the fluid simulation is higher than the water level at the pipe network node, the surface water flows into the pipe network through the storm drain (grate), becoming the inflow at the SWMM node. The calculation expression is as follows:

[0083]

[0084]

[0085]

[0086] in, For the first Time step, surface grid (the first) Line 1 (Column) Merging into SWMM network nodes Inflow; The flow coefficient of the rainwater grate (range 0.6-0.8); The effective flow area of ​​the rainwater inlet grate; It is the acceleration due to gravity; This represents the absolute water level difference between the surface grid and the SWMM network nodes (only positive values ​​are considered). The available water volume (current water accumulation volume) of the surface grid. For the first Time step, SWMM network node The absolute water level; For the surface grid (the first Line 1 The topographic elevation of (a series of columns);

[0087] S5.3. Data interaction uses Socket communication in JSON format, including timestamp, grid ID / node ID, and flow / water level values;

[0088] S6. Construct a real-time multi-channel volume rendering pipeline to perform real-time simulation rendering of surface runoff, including a time synchronization module, a terrain and water level initialization buffer module, an outflow calculation buffer module, a water level correction buffer module, an outflow optimization buffer module, a volume rendering module, and an SWMM data interaction interface module. Together, they complete the real-time calculation and real-time volume rendering display of various data in step S2 (surface water production), S3 (water volume in the grid and water flow between grids), S4 (node ​​water level and pipe flow in the SWMM network), and S5 (inflow / outflow of the SWMM network).

[0089] S6.1. Construct a time synchronization module to synchronize the starting time of rainfall, SWMM simulation, and PEHDCM simulation , and the starting time of each step calculation is calibrated;

[0090] S6.2. In the Cesium three-dimensional engine, write a terrain and water level initialization buffer module using the GLSL shader language, integrate terrain, rainfall, SWMM overflow, and other data sources to generate an initial water level texture. The input data includes: terrain height map texture (single channel, storing normalized terrain elevation), terrain real range texture (storing the minimum and maximum values of the real elevation range for normalization conversion), rainfall texture (single channel, storing the water level data converted from the real rainfall intensity at the current time step), infiltration amount texture (single channel, storing the infiltration amount at the current time step), SWMM overflow texture (single channel, storing the real overflow amount from the SWMM node to the ground at the previous time step), and the previous step water level texture (single channel, storing the normalized ground water level at the previous time step). Superimpose the texture data to obtain the initial water level texture;

[0091] S6.3. In the Cesium three-dimensional engine, write a flow calculation buffer module using the GLSL shader language, based on the initial water level texture, calculate the outflow of the grid in four directions, and generate a flow texture. The input data includes the initial normalized water level texture calculated in S6.2, the normalized terrain texture (used for calculating the slope), and the slope texture (pre-calculated normalized slope in four directions). The calculation process is as follows: for each grid, sample the normalized water level and the four domain water levels, calculate the normalized water level difference; based on the slope texture and the water level difference, calculate the normalized outflow in four directions; generate the outflow texture in four directions (four-channel texture, each channel corresponds to the normalized outflow in one direction) according to the total flow constraint in S3.2.5;

[0092] S6.4. In the Cesium three-dimensional engine, write a water level correction buffer module using the GLSL shader language, combine the inflow and SWMM inflow to correct the water level texture and ensure water conservation. The input data includes the outflow texture calculated in S6.3, the SWMM node water level texture, and the node-grid association texture (marking the SWMM node ID associated with each grid). The calculation process is as follows: for each grid, accumulate the inflow from the four neighboring domains, generate a grid interaction inflow texture; preliminarily correct the normalized water level; calculate the ground inflow to the SWMM (the calculation method in step S5.2); output the corrected normalized water level texture;

[0093] S6.5. In the Cesium three-dimensional engine, a flow optimization buffer module is written using the GLSL shader language, based on the corrected water level, to optimize the outflow, generate the final flow texture, and use it for the next iteration. The input data includes: the corrected water level texture in S6.4; the slope texture in S6.3. The calculation process: recalculate the four-direction normalized water level difference based on the corrected water level texture; optimize the outflow to ensure matching with the corrected water level; output the optimized normalized flow texture;

[0094] S6.6. In the Cesium three-dimensional engine, a volume rendering module is written using the GLSL shader language to convert the normalized water level to real values and render fluid effects in combination with the terrain. The input data includes: the corrected normalized water level texture, the normalized terrain texture, and the terrain real range texture. The rendering process: convert the corrected normalized water level texture and the terrain height map texture to real values; calculate the real water depth for adjusting the water surface color (e.g., deep water is blue, shallow water is transparent); combine lighting, normal calculation (based on terrain slope) to render water surface light and shadow effects, and output the fluid visualization picture, as shown in Figure 2

[0095] S6.7. A POST request interaction interface service is written using C++ to build SWMM data, and a corresponding calling function is written using JavaScript in Cesium to pass the ground inflow data to SWMM and receive the SWMM overflow data (for the next iteration). The input data is the real inflow texture to SWMM, and the output data is the current step overflow texture calculated by SWMM, which is used for the next iteration;

[0096] S6.8. Write a loop execution statement to iteratively execute S6.1, S6.2, S6.3, S6.5, S6.6, and S6.7, as shown in the execution flow Figure 3

[0097] Embodiment 2

[0098] A system for constructing an efficient and lightweight urban waterlogging real-time simulation model, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the efficient and lightweight urban waterlogging real-time simulation model construction method of embodiment 1 when executed.​​

Claims

1. A method for constructing an efficient and lightweight real-time simulation model of urban flooding, characterized in that, The process includes the following steps: S1. Collecting data from the modeling area, including rainfall time-series data, underlying digital surface model (DSM) raster data, soil distribution raster data, and underground drainage network data; S2. Creating a topographic height map texture using the underlying digital surface model (DSM) raster data from step S1; converting and calculating the actual surface water level increment using the rainfall time-series data from step S1, and creating rainfall-based water level increment texture data; creating infiltration rate texture data using the soil distribution raster data from step S1, and generating actual surface water yield based on the rainfall water level increment and infiltration rate; S3. Constructing a Physical Parameter Enhanced Height Difference-Flow Conservation Surface Hydrodynamic Model (PEHDCM); S4. Reconstruct the SWMM network model, retaining only the transport components. Construct the SWMM network model using the underground drainage network data from step S1, and create a grid association mapping table between network model nodes and topographic height map textures based on the topographic height map textures from step S2. This table is used to record and identify the grids where network nodes are located. S5. The actual surface water production in step S2 is used as the inflow data of the SWMM network model node, and the outflow of the SWMM network model node is used as the water level increment source term in the surface fluid simulation process. A two-way data interaction is completed every Δt seconds, where Δt is the simulation time step of the SWMM model, realizing the two-way dynamic coupling between the network and the surface water flow, and calculating the surface water level and flow direction in real time; S6. A real-time multi-channel volume rendering pipeline is constructed to perform real-time simulation rendering of surface runoff fluid. The rendering pipeline includes a time synchronization module, a terrain and water level initialization buffer module, an outflow calculation buffer module, a water level correction buffer module, an outflow optimization buffer module, a volume rendering module, and an SWMM data interaction interface module, which jointly complete real-time calculation and real-time volume rendering display.

2. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 1, characterized in that, In step S1: The spatial resolution of the underlying digital surface model (DSM) raster data is 2 meters. The soil distribution raster data is obtained from a soil database or through remote sensing image inversion and interpretation, with a spatial resolution of 2 meters; the underground rainwater pipe network data includes attribute data of node elements and connection elements; the rainfall time series data is in millimeters per hour and is generated by collecting data or using the Chicago rainstorm intensity formula.

3. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 2, characterized in that, The specific implementation of step S2 includes: S2.

1. Create terrain height map texture data based on the underlying digital surface model (DSM) raster data, with pixel size consistent with the DSM spatial resolution, which is 2×2 meters; S2.

2. Based on the soil distribution raster data, calculate the dynamic permeability of each grid to obtain infiltration texture data; S2.

3. Based on the rainfall time series data, calculate and convert it into the actual water level increment data of the modeling area to obtain the rainfall-based water level increment texture data.

4. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 3, characterized in that, The specific implementation method of step S3 includes the following steps: S3.

1. Based on the topographic height map texture, the surface area is discretized using a regular grid of equal resolution. The grid cell state variables include water level and outflow in four directions. S3.

2. Discrete calculations are performed based on the Physical Parameter Enhanced Height Difference-Flow Conservation Model (PEHDCM), including velocity derivation, flow correction, viscous smoothing, water level update, and physical constraint steps. S3.2.

1. Derivation of flow velocity based on water level gradient and gravity; S3.2.

2. Achieve viscous smoothing of water body by weighted averaging of flow velocities from adjacent grids; S3.2.

3. Integrate height difference, flow velocity, and gravity for flow rate correction; S3.2.

4. Perform water level update calculations including velocity and viscosity corrections; S3.2.

5. Apply physical constraints, including water conservation constraints and flow velocity constraints.

5. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 4, characterized in that, The specific implementation method of step S4 includes the following steps: S4.

1. Reconstruct the SWMM network model, retaining only the transport components, and construct the SWMM stormwater network hydraulic model; S4.

2. Based on the real spatial coordinates of each grid in the terrain height map texture data, overlay the SWMM network model topology data to create a grid association mapping table between network model nodes and terrain height map textures.

6. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 5, characterized in that, The specific implementation of the bidirectional dynamic coupling in step S5 includes: S5.

1. Forward coupling: The overflow flow of the pipe network nodes calculated by the SWMM stormwater pipe network hydraulic model is used as the source term for the surface fluid simulation and superimposed on the corresponding surface grid; S5.

2. Reverse coupling: When the surface water level is higher than the water level of the pipe network nodes, the inflow of surface water into the pipe network is calculated; S5.

3. Data interaction adopts Socket communication, with JSON format, including timestamp, grid ID / node ID, flow / water level values.

7. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 6, characterized in that, The construction of the multi-channel volume rendering pipeline in step S6 includes: S6.

1. Constructing a time synchronization module to synchronize the start time of each simulation and the start time of each calculation step; S6.

2. Constructing a terrain and water level initialization buffer module to integrate various data sources and generate initial water level textures; S6.

3. Constructing a flow calculation buffer module to calculate the outflow in four directions based on the initial water level texture; S6.

4. Constructing a water level correction buffer module to correct the water level texture by combining the inflow and the inflow of the SWMM stormwater network hydraulic model; S6.

5. Constructing a flow optimization buffer module to optimize the outflow based on the corrected water level; S6.

6. Constructing a volume rendering module to convert the normalized water level into a real value and render the fluid effect; S6.

7. Constructing a data interaction interface module with the SWMM stormwater network hydraulic model to realize data transmission between the ground and the network.

8. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 7, characterized in that, The normalization process in the multi-channel volume rendering pipeline converts physical quantities into a uniform numerical range, and the conversion expression is as follows: in, Normalized water level; This represents the actual surface water level. and These represent the minimum and maximum values ​​of the actual terrain elevation span of the modeling area.

9. The method for constructing an efficient and lightweight real-time urban flooding simulation model according to claim 8, characterized in that, The textures used in the multi-channel volume rendering pipeline are two-dimensional pixel arrays that carry specific data, serving as data containers for GPU computation.

10. A system for constructing an efficient and lightweight real-time simulation model of urban flooding, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed, implements the steps of the efficient and lightweight real-time simulation model construction method for urban flooding as described in any one of claims 1-9.

Images