Sponge city rainwater garden storm runoff infiltration simulation method and system

Abstract

This invention discloses a method and system for simulating stormwater runoff infiltration and storage in rain gardens for sponge cities. First, using a centimeter-resolution three-dimensional topographic model of the target catchment area, the layered structure and geotechnical parameters of LID (Light Infiltration Device) facilities, and minute-by-minute rainfall intensity-duration curves of the design storm as input, an improved Green-Ampt infiltration model considering the layered media and surface water storage driving effects is constructed. This model is integrated with a two-dimensional shallow water equation surface runoff model driven by centimeter-resolution micro-topography through a two-way temporal coupling mechanism, achieving minute-by-minute dynamic simulation of the entire design storm process. This fully reproduces the entire hydrological process of surface runoff collection, LID facility infiltration and storage, and overflow. This invention significantly improves the accuracy and reliability of stormwater process simulation, and can simultaneously output spatiotemporal dynamic visualization results of runoff, quantitative hydrological indicators, and compliance assessment results of sponge city design specifications, providing reliable technical support for the refined design and effect verification of LID facilities.

Description

Technical Field

[0001] This invention relates to the field of sponge city design technology, specifically to a method and system for simulating stormwater runoff infiltration and storage in sponge city rain gardens. Background Technology

[0002] In the construction of sponge cities, rain gardens, sunken green spaces and other LID facilities are the core carriers for realizing the retention and infiltration of urban storm runoff, reducing the peak drainage of pipe networks, and preventing urban flooding. Current design specifications have clear and mandatory requirements for the runoff peak reduction rate of LID facilities under the set design storm.

[0003] In current engineering practice, the mainstream approach to designing and evaluating the effectiveness of LID (Low Impact Drainage) facilities is to use the runoff coefficient method for static empirical estimation. This method cannot predict the start time of overflow, the temporal distribution of overflow volume, or verify its compatibility with the drainage capacity of the downstream municipal pipe network. This can easily lead to the actual effect of the design scheme failing to meet the standards and poses a safety hazard for flood prevention.

[0004] Existing hydrological runoff simulation methods mostly employ generalized calculations at the catchment scale, simplifying the depression depth of LID facilities to a single uniform value. This ignores the precise control of centimeter-level micro-topography on runoff confluence paths, confluence times, and overflow boundary elevations, making it difficult to meet the requirements of refined design in terms of simulation accuracy. At the same time, the infiltration process often uses the traditional Green-Ampt model, which does not consider the infiltration driving effect of the layered medium structure of LID facilities, surface water depth, and interlayer percolation effect. This results in insufficient accuracy in calculating infiltration volume. Furthermore, the infiltration model and surface runoff model are mostly unidirectionally coupled, failing to achieve real-time bidirectional feedback between infiltration and runoff throughout the entire rainfall process, leading to poor reliability of simulation results.

[0005] In addition, the output of existing simulation technologies are mostly static final value indicators such as total runoff and peak reduction rate, lacking minute-by-minute time-series simulation and three-dimensional visualization of the entire design storm process. They have not formed a compliance verification closed loop that directly matches the design specifications for sponge cities, making it difficult to support the refined design and rapid optimization of LID facilities. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention discloses a method and system for simulating stormwater runoff infiltration in sponge city rain gardens, in order to solve the problems mentioned in the background.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for simulating stormwater runoff infiltration in a sponge city rain garden, comprising the following steps:

[0008] S1. Obtain a centimeter-resolution 3D terrain model of the target catchment area, hierarchical structure data of LID facilities, geotechnical engineering parameters, minute-by-minute rainfall intensity-duration curves of the design storm, and construct an integrated simulation base dataset.

[0009] S2. Based on the layered structure of LID facilities, an improved Green-Ampt infiltration model is constructed to calculate the real-time infiltration of LID facilities and surrounding underlying surfaces during rainfall.

[0010] S3. Using high-precision micro-topographic elevation data from a centimeter-level three-dimensional terrain model, a two-dimensional shallow water equation-driven surface runoff calculation model is constructed to simulate the spatiotemporal convergence and evolution of surface runoff within the catchment area.

[0011] S4. The improved Green-Ampt infiltration model and the two-dimensional shallow water equation model are integrated through a two-way time-series coupling mechanism to perform minute-by-minute dynamic simulation of the entire process of the design storm, covering the entire hydrological process of surface runoff collection, water storage and infiltration of LID facilities and overflow.

[0012] S5. Based on the coupled simulation results, output the dynamic visualization results of the spatiotemporal distribution of runoff throughout the entire rainstorm process, the hydrological quantitative indicators of LID facilities, and the compliance assessment results with the sponge city drainage design specifications.

[0013] Preferably, step S1 specifically includes:

[0014] S11. Obtain a three-dimensional terrain model of the target catchment area with centimeter-level resolution through three-dimensional terrain scanning. The model includes micro-topographic elevation data of the entire catchment area, spatial boundaries of LID facilities, elevation data of each position of the concave edge, spatial location and elevation data of drainage outlets, and canopy interception parameters corresponding to plant configuration.

[0015] S12. Collect layered structural data of LID facilities, including the layer thickness data of planting soil layer, gravel cushion layer and drainage layer; collect corresponding geotechnical engineering parameters, including saturated permeability coefficient, field water holding capacity, saturated water content and wetting front matrix suction head of each layer medium.

[0016] S13. Obtain the minute-by-minute rainfall intensity-duration curve of the design storm for the corresponding design return period, and complete the construction of the integrated simulation basic dataset.

[0017] Preferably, in step S2, the construction of the improved Green-Ampt infiltration model and the calculation of real-time infiltration volume specifically include:

[0018] For the layered media structure of LID facilities, an improved Green-Ampt infiltration model is constructed, considering surface water depth-driven infiltration, stratified wetting front propagation, and interlayer percolation effect. The infiltration rate f(t) at any calculation time t is calculated using the following formula:

[0019] ;

[0020] In the formula: Ks h is the saturated permeability coefficient of the infiltration layer where the current wetting front is located. w To calculate the real-time water depth within the LID facility at any given time; h s ψ is the thickness of the saturated soil layer above the infiltration layer where the current wetting front is located; f The matrix suction head at the wetting front of the current infiltration medium; z f (t) represents the propagation depth of the wetting front within the current infiltration layer at time t; when the wetting front penetrates the current infiltration layer and enters the next layer, the K value of the corresponding layer is updated. s With ψ f The calculation parameters are adjusted to account for the interlayer percolation effect caused by the difference in permeability coefficients between upper and lower layers, and the infiltration rate calculation results are corrected. For the surrounding underlying surface outside the LID facility, the real-time infiltration rate is calculated using a single-layer Green-Ampt model with corresponding underlying surface medium parameters. Based on the infiltration rate at each time step, the cumulative infiltration rate for the corresponding time period is obtained by integration.

[0021] Preferably, in step S3, the construction and solution of the surface runoff calculation model driven by the two-dimensional shallow water equation specifically includes:

[0022] Based on a 3D terrain model with centimeter-level resolution, the target catchment area is meshed. The bottom elevation of each mesh cell is directly taken from the corresponding location micro-topographic elevation data of the 3D terrain model. The conservation form of the two-dimensional shallow water equations is then constructed as follows:

[0023] ;

[0024] In the formula: U is the vector of conserved variables. , where h is the real-time water depth within the grid cell, and u and v are the cross-sectional average velocity components in the x and y directions, respectively;

[0025] F(U) is the flux vector in the x-direction. , where g is the acceleration due to gravity;

[0026] G(U) is the flux vector in the y-direction. ;

[0027] S(U) is the source term vector, which includes the bottom slope source term, friction source term, rainfall source term, and infiltration sink term. The bottom slope source term is calculated based on the micro-topographic elevation gradient of the grid cell, the friction source term is calculated using the Manning formula, the rainfall source term is the rainfall intensity of the corresponding time period, and the infiltration sink term is the real-time infiltration rate of the corresponding grid cell obtained in step S2. The finite volume method is used to discretize and solve the governing equations to obtain the time-by-time water depth and velocity distribution of each grid cell in the catchment area, simulating the spatiotemporal collection and evolution process of surface runoff.

[0028] Preferably, the bidirectional temporal coupling mechanism in step S4 adopts a unified minute-by-minute calculation time step that matches the design rainstorm time series, and executes the following bidirectional coupling calculation loop within each time step:

[0029] The first step is to calculate the real-time infiltration and cumulative infiltration of each grid cell in the current time step based on the rainfall intensity at the current time step and the global surface water depth distribution obtained from the previous time step, using the improved Green-Ampt infiltration model. The cumulative infiltration is then used as the infiltration sink term in the governing equation of the two-dimensional shallow water equation.

[0030] The second step is to substitute the updated infiltration sink term into the two-dimensional shallow water equation governing equation to solve for the water depth distribution, flow velocity distribution, and real-time water storage depth within the LID facility range of each grid cell in the current time step.

[0031] The third step is to feed back the real-time water depth within the LID facility to the improved Green-Ampt infiltration model, updating the h value used to calculate the infiltration rate for the next time step. w Parameters are used to complete a single bidirectional coupling calculation loop;

[0032] Repeat the above bidirectional coupled calculation loop until the minute-by-minute dynamic simulation of the entire design rainstorm process is completed.

[0033] Preferably, in the second step of the bidirectional coupled calculation loop, the overflow process simulation of the LID facility is performed synchronously, specifically including:

[0034] Based on a centimeter-level 3D terrain model, the overflow elevation threshold of each grid cell at the recessed edge of the LID facility is pre-obtained;

[0035] After obtaining the water depth distribution at each time step, the real-time water depth of each grid cell at the edge of the LID facility is compared with the overflow elevation threshold at the corresponding location. When the real-time water depth exceeds the overflow elevation threshold, it is determined that an overflow has occurred at that location. The overflow flow rate of the grid cell is calculated and incorporated into the flux calculation of the two-dimensional shallow water equation. The overflow start time, the overflow flow rate process line per minute, and the cumulative overflow flow rate are output synchronously.

[0036] Preferably, the dynamic visualization results of runoff spatiotemporal distribution in step S5 specifically include:

[0037] The animation of water depth distribution throughout the entire rainstorm process, built on a centimeter-level 3D terrain model, uses continuous color levels to represent the real-time water depth values ​​of different grid units. Simultaneously, runoff direction vectors, overflow locations, and overflow volume indicators are overlaid to fully present the spatiotemporal evolution of the entire process of surface runoff collection, LID facility water storage, and overflow.

[0038] Preferably, the hydrological quantitative indicators and compliance assessment in step S5 specifically include:

[0039] The hydrological quantitative indicators include the time history curves of the water storage depth, runoff infiltration, overflow, total runoff and peak runoff at the outlet of the catchment area of ​​the LID facility.

[0040] The compliance assessment compares the peak runoff reduction rate calculated by simulation with the threshold corresponding to the sponge city design code, outputs the judgment result of whether the current LID facility design scheme meets the code requirements, and outputs the direction of design parameter optimization when the code requirements are not met.

[0041] This invention also provides a dynamic simulation system for stormwater runoff infiltration in sponge city rain gardens, comprising the following modules connected in sequence via communication:

[0042] The data input and preprocessing module is used to acquire a centimeter-resolution three-dimensional terrain model of the target catchment area, layered structural data of LID facilities, geotechnical engineering parameters, and minute-by-minute rainfall intensity-duration curves of the design storm, and to build an integrated simulation base dataset.

[0043] The infiltration model construction module is used to build an improved Green-Ampt infiltration model based on the layered structure of LID facilities, and to calculate the real-time infiltration of LID facilities and surrounding underlying surfaces during rainfall.

[0044] The surface runoff simulation module is used to construct a two-dimensional shallow water equation-driven surface runoff calculation model using high-precision micro-topographic elevation data from a centimeter-level three-dimensional terrain model, simulating the spatiotemporal convergence and evolution of surface runoff within the catchment area.

[0045] The two-way coupled simulation module is used to integrate the improved Green-Ampt infiltration model with the two-dimensional shallow water equation model through a two-way time-series coupling mechanism to perform minute-by-minute dynamic simulation of the entire process of the design storm, covering the entire hydrological process of surface runoff collection, water storage and infiltration of LID facilities and overflow.

[0046] The results output and compliance verification module is used to output dynamic visualization results of the spatiotemporal distribution of runoff throughout the entire rainstorm process, hydrological quantitative indicators of LID facilities, and compliance assessment results with sponge city drainage design specifications, based on the coupled simulation results.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] 1. This invention improves the Green-Ampt infiltration model by considering the LID layered medium structure and the surface water storage driving effect, and constructs a two-way time-series coupling mechanism with the two-dimensional shallow water equation. This enables real-time two-way feedback between the infiltration process and the evolution of surface runoff throughout the entire rainfall process. It can accurately reproduce the complete hydrological dynamics of runoff collection, storage and infiltration, and overflow throughout the entire rainstorm cycle, and accurately predict the overflow start time and overflow volume time-series distribution. This fundamentally solves the problem of result distortion in traditional static estimation methods.

[0049] 2. This invention uses a three-dimensional terrain model with centimeter-level resolution to drive the solution of two-dimensional shallow water equations, refining the simulation accuracy from the traditional catchment area scale to the centimeter-level micro-topography scale. It accurately depicts the control effect of micro-topographic undulations on runoff confluence paths and confluence times, as well as the true overflow elevation thresholds at various locations on the concave edge of LID facilities. This completely avoids the overflow simulation distortion problem caused by the generalization of uniform concave depth, and provides high-precision spatial simulation support for the refined design of LID facilities.

[0050] 3. This invention realizes minute-by-minute time-series simulation calculation of the entire process of a design storm, and simultaneously outputs three-dimensional dynamic visualization results of runoff spatiotemporal distribution, quantitative indicators of the entire hydrological process of LID facilities, and compliance assessment results that directly match the current sponge city design specifications. It forms a complete technical closed loop of simulation calculation - effect quantification - compliance judgment - optimization guidance, which significantly reduces the design threshold of LID facilities, effectively avoids the risk of substandard schemes caused by experience-based design, and can directly support the standardized design work of landscape engineers. Attached Figure Description

[0051] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0052] In the attached diagram:

[0053] Figure 1 This is a flowchart of a method for simulating stormwater runoff infiltration in a sponge city rain garden according to the present invention;

[0054] Figure 2 This is a structural diagram of a rain garden storm runoff infiltration simulation system for sponge cities, according to the present invention. Detailed Implementation

[0055] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0056] Example: The implementation platform of this example is an engineering computing workstation equipped with a 64-bit Windows operating system and 32G of RAM. The simulation calculation adopts a fixed time step of 1 minute that matches the designed rainstorm timing, and follows the method flow and system architecture defined in the claims of this invention throughout.

[0057] like Figure 1 As shown, this invention provides a method for simulating stormwater runoff infiltration in a sponge city rain garden, the specific implementation steps of which include:

[0058] S1. Simulate the construction of the basic dataset. The specific implementation process is as follows:

[0059] S11. Acquisition of a centimeter-level 3D terrain model: A ground-based 3D laser scanning device is used to perform a full-area scan of the target catchment area, acquiring a 2cm×2cm resolution centimeter-level 3D terrain model. The model covers the entire target catchment area and includes the following core data:

[0060] 1) Basic parameters of the catchment area: The total area of ​​the catchment area is 1200㎡, of which the rain garden LID facility occupies an area of ​​180㎡, located in the low-lying part of the catchment area, accounting for 15% of the catchment area;

[0061] 2) Micro-topography and boundary data: point-by-point micro-topographic elevation data of the entire catchment area, closed spatial boundary of the rain garden, point-by-point measured elevation data of 128 sampling points on the concave edge of the rain garden, spatial coordinates and bottom elevation of the drainage outlet at the end of the rain garden;

[0062] 3) Vegetation parameters: The rain garden plant configuration is a combination of native herbs and flowering shrubs, and the maximum canopy interception capacity is set to 0.8mm.

[0063] S12. Layered Structure and Geotechnical Engineering Parameter Acquisition of LID Facility: The rain garden in this embodiment adopts the standard vertical layered structure of sponge city, which consists of planting soil layer, gravel cushion layer and perforated drainage layer from top to bottom. The structural parameters of each layer and the measured geotechnical engineering parameters are detailed in Table 1.

[0064] S13. Acquisition of design storm time series data: Using the current storm intensity formula of the project location, the total rainfall of the 5-year return period 30-minute design storm is calculated to be 34.2 mm in this embodiment. The Chicago rain pattern is used for rain pattern allocation, and the peak rainfall coefficient is taken as 0.4 to obtain the minute-by-minute rainfall intensity-duration curve. The rainfall peak occurs at the 12th minute, and the peak rainfall intensity is 2.87 mm / min. The above data is standardized and integrated to construct an integrated simulation basic dataset.

[0065] Table 1. Layered Structure and Geotechnical Engineering Parameters of Rain Garden LID Facilities

[0066] Layered name Layered thickness (mm) Saturated permeability coefficient (m / s) Saturated water content Field water holding capacity Wetting front matrix suction head (m) Manning coefficient Planting soil layer 300 5 x 10⁻ 5 ]] 0.45 0.22 0.16 0.24 Gravel cushion layer 200 8×10⁻³ 0.38 0.12 0.03 0.18 Perforated drainage layer 100 2×10⁻² 0.40 0.05 0.01 0.10 Catchment area ordinary green land - 5 x 10⁻ 5 ]]> 0.45 0.22 0.16 0.15

[0067] S2. The construction of the improved Green-Ampt infiltration model and the calculation of real-time infiltration volume are implemented as follows:

[0068] For the layered media structure of the rain garden in this embodiment, an improved Green-Ampt infiltration model is constructed, which considers surface water depth driving, layered wetting front propagation, and interlayer water retention effect, to replace the traditional single-layer Green-Ampt model that does not consider the layered structure and the influence of surface water storage. The specific calculation rules are as follows: The infiltration rate f(t) at any calculation time t is calculated using the following formula:

[0069] ,

[0070] In the formula: K s h represents the saturated permeability coefficient of the infiltration layer where the current wetting front is located; the parameter values ​​are directly taken from Table 1. w To calculate the real-time water depth within the rain garden grid cell at a given time, the result is obtained from the solution of the two-dimensional shallow water equation at the previous coupling step; h s ψ is the thickness of the saturated soil layer above the infiltration layer where the current wetting front is located; f The matrix suction head at the wetting front of the current infiltration medium is taken directly from Table 1; z f (t) represents the depth of the wetting front within the current infiltration layer at time t.

[0071] In this embodiment, the correction rule for layer penetration and interlayer water retention effect is as follows: when the wetting front advances to a depth z f (t) When the total thickness of the planting soil layer reaches 300mm, it is determined that the wetting front has penetrated the current layer, and the calculation parameters of the crushed stone cushion layer are automatically switched.

[0072] Since the saturated permeability coefficient of the lower gravel cushion layer in this embodiment is much greater than that of the upper planting soil layer, there is no water retention phenomenon between layers, so there is no need to reduce the infiltration rate; if the permeability coefficient of the lower medium is less than that of the upper layer, the infiltration rate is reduced and corrected according to the ratio of the permeability coefficients of the upper and lower layers to restore the true hydrological effect of water retention between layers.

[0073] For the surrounding green space grid outside the rain garden, the real-time infiltration rate is calculated using the single-layer Green-Ampt model corresponding to the ordinary green space in Table 1. Based on the infiltration rate calculation results with a minute-by-minute time step, the cumulative infiltration rate for each grid unit in the corresponding time period is obtained by time integration, providing infiltration sink terms for subsequent two-dimensional shallow water equation calculation.

[0074] S3. Construction of a two-dimensional shallow water equation-driven surface runoff calculation model, the specific implementation process of which is as follows:

[0075] Based on the 2cm resolution 3D terrain model obtained by S11, in order to balance computational accuracy and efficiency, the target catchment area is divided into a 10cm×10cm structured grid, with a total of 120,000 computational grid cells. The bottom elevation of each grid cell is directly extracted from the 3D terrain model by interpolation, which fully preserves the centimeter-level micro-topographic undulation features of the catchment area, replacing the uniform elevation generalization method of the catchment area and LID facilities in the traditional model.

[0076] Based on the above grid system, the conservation form of the two-dimensional shallow water equations is constructed to simulate the spatiotemporal accumulation and evolution of surface runoff. The governing equations are as follows:

[0077] ,

[0078] U is a vector of conserved variables. Where h is the real-time water depth within the grid cell, and u and v are the cross-sectional average velocity components in the x and y directions, respectively.

[0079] F(U) is the flux vector in the x-direction. Where g is the acceleration due to gravity. 2

[0080] G(U) is the flux vector in the y-direction.

[0081] S(U) is the source term vector, which includes the bottom slope source term, friction source term, rainfall source term, and infiltration sink term, where:

[0082] This embodiment uses the finite volume method to discretize and solve the governing equations, employs the Roe scheme to calculate the flux at the grid interface, and uses a semi-implicit scheme to process the source terms, avoiding numerical oscillations caused by alternating wet and dry interfaces and ensuring computational stability. Finally, it obtains the minute-by-minute distribution of water depth and velocity in each grid cell within the catchment area, fully simulating the spatiotemporal collection and evolution of surface runoff.

[0083] The complete expression for the source term vector S(U) is:

[0084] ,

[0085] In the formula:

[0086] The first row corresponds to the source and sink terms of the continuity equation, where i is the minute-by-minute rainfall intensity corresponding to the current time step, with the dimension uniformly in m / s; f is the real-time infiltration rate of the corresponding grid cell calculated in step S2, with the dimension uniformly in m / s.

[0087] The second and third rows correspond to the source terms of the momentum equation, where:

[0088] Bottom slope source term: , The bottom slope source terms are in the x and y directions, respectively, and z b The bottom elevation of the grid cells is directly extracted from the centimeter-level 3D terrain model. , The elevation gradient of the micro-topography in adjacent grid cells is used to fully restore the control effect of micro-topography on runoff direction;

[0089] Friction source term: , The friction source terms in the x and y directions are respectively calculated using the Manning formula, and the specific expression is as follows:

[0090] , ;

[0091] In the formula, n is the Manning coefficient of the underlying surface of the corresponding grid cell, and the value is directly taken from Table 1;

[0092] Rainfall source term: the minute-by-minute rainfall intensity i corresponding to the current time step, uniformly converted to m / s for calculation;

[0093] Infiltration sink term: The real-time infiltration rate f of the corresponding grid cell obtained from step S2 is uniformly converted to m / s for calculation; when the water depth h of the grid cell is less than the product of the infiltration rate and the time step, the infiltration rate is taken as h / Δt to avoid abnormal negative water depth values.

[0094] Specific processing rules:

[0095] Interface flux calculation: The numerical flux of the mesh interface is calculated using the Roe format. For dry-wet interfaces (one side of the mesh has a water depth less than the dry water depth threshold and the other side is a wet mesh), the modified Roe format is used for flux calculation to avoid numerical oscillations and non-physical negative fluxes caused by alternating dry and wet conditions.

[0096] Source term discretization: The bottom slope source term and friction source term are discretized using a semi-implicit scheme. The bottom slope source term is discretized using the hydrostatic pressure reconstruction method to ensure computational harmony, while the friction source term is implicitly discretized to adapt to the nonlinear characteristics of the Manning formula and improve computational stability.

[0097] Initial and boundary conditions: The initial conditions are set as follows: the initial water depth and initial flow velocity are 0 for the entire grid. The boundary conditions include three types: ① The land boundary of the catchment area adopts a non-slip solid wall closed boundary, and the normal flow velocity is set to 0; ② The overflow boundary of the rain garden adopts the broad-crested weir formula to calculate the overflow flow rate and incorporates it into the interface flux calculation; ③ The drainage outlet at the end of the rain garden adopts a free outflow boundary condition.

[0098] Stability and convergence control: The time step is controlled by CFL conditions, and the Coulomb number is set to no more than 0.5, matching the 1-minute macroscopic calculation step of this scheme; the dry water depth threshold is set to 1×10⁻ 6 m, when the water depth of a grid cell is less than this threshold, it is considered a dry grid and does not participate in the momentum equation solution; the convergence criterion for a single time step is that the L2 norm of the residual of the continuity equation is less than 1×10⁻ 6 .

[0099] Through the above discrete solution process, the water depth and flow velocity distribution of each grid cell in the catchment area are obtained minute by minute throughout the entire design rainstorm process, which fully simulates the spatiotemporal convergence and evolution process of surface runoff.

[0100] S4. Two-way temporal coupling of infiltration-runoff model and dynamic simulation of the entire hydrological process, the specific implementation process is as follows:

[0101] The improved Green-Ampt infiltration model constructed in S2 and the two-dimensional shallow water equation model constructed in S3 are integrated through a two-way time-series coupling mechanism. A unified calculation time step of 1 minute, which is completely matched with the design storm time series, is adopted. The following closed-loop two-way coupled calculation cycle is executed within each time step:

[0102] The first step is to update the infiltration rate: Based on the rainfall intensity at the current time step and the global surface water depth distribution obtained from the previous time step, the real-time infiltration rate and cumulative infiltration of each grid cell in the current time step are calculated using the improved Green-Ampt infiltration model. The cumulative infiltration is then used as the infiltration sink term in the governing equation of the two-dimensional shallow water equation.

[0103] The second step is to solve for the runoff and overflow processes: Substitute the updated infiltration sink term into the governing equations of the two-dimensional shallow water equations to obtain the water depth distribution, velocity distribution, and real-time water storage depth within the rain garden area for each grid cell in the current time step; simultaneously, the overflow process simulation is performed, with the following specific rules:

[0104] Based on the overflow elevation threshold of each grid cell at the concave edge of the rain garden pre-extracted from the 3D terrain model, the real-time water depth + bottom elevation of the edge grid cell is compared with the overflow elevation threshold at the corresponding location. When the calculated value exceeds the overflow elevation threshold, it is determined that an overflow has occurred at that location. The overflow flow rate of the grid cell is calculated according to the broad-crested weir formula, and the overflow flow rate is incorporated into the flux calculation of the 2D shallow water equation. The overflow start time, minute-by-minute overflow flow rate and cumulative overflow flow rate are recorded simultaneously.

[0105] The third step is to provide infiltration parameter feedback: This involves monitoring the real-time water depth h within the rain garden. w Feedback is fed back to the improved Green-Ampt infiltration model to update the h of the infiltration rate calculation for the next time step. wThe parameters are used to complete a single bidirectional coupling calculation loop.

[0106] Repeat the above bidirectional coupling calculation loop until the minute-by-minute dynamic simulation of the entire 30-minute design storm process is completed, fully covering the entire hydrological process of surface runoff collection, LID facility water storage and infiltration, and overflow discharge.

[0107] This embodiment achieves real-time bidirectional feedback between the infiltration process and the surface runoff process through a two-way coupling mechanism, which solves the defect of the traditional one-way coupling model that cannot reproduce the real hydrological feedback mechanism of water depth affecting infiltration rate and infiltration rate affecting runoff depth.

[0108] S5. Simulation results output and compliance assessment, the specific implementation process is as follows:

[0109] Based on the full-process coupled simulation results of S4, three core outputs are generated, forming a complete closed loop of simulation calculation, effect quantification, and compliance determination:

[0110] The first category is dynamic visualization results of runoff spatiotemporal distribution: Based on a centimeter-level 3D terrain model, a minute-by-minute water depth distribution rendering animation of the entire process of the designed rainstorm is generated. The real-time water depth values ​​of the grid cells from 0 to 150 mm are represented by a continuous blue-red gradient color scale. The runoff direction vector, overflow location and real-time overflow volume are simultaneously superimposed to fully present the spatiotemporal evolution process of surface runoff collection, rain garden water storage and overflow discharge. This solves the problem that traditional models can only output static values ​​and cannot intuitively reproduce the dynamic process of rain and flood.

[0111] The second category is quantitative hydrological indicators for LID facilities, including time-history curves of rain garden water storage depth, runoff infiltration, overflow, total runoff and peak runoff at the catchment outlet.

[0112] The core indicators simulated in this embodiment are: the maximum water storage depth of the rain garden is 280mm (occurring at the 18th minute of rainfall), the overflow starts at the 14th minute of rainfall, and the cumulative overflow is 2.1m³; the peak runoff flow at the outlet of the catchment area without LID facilities is 0.072m³ / s, and the peak runoff flow after the implementation of LID facilities in this scheme is 0.027m³ / s.

[0113] The third category is the compliance assessment results of the sponge city design specifications: Based on the peak flow data obtained from the simulation, the runoff peak reduction rate of this scheme is calculated to be 62.5%. Compared with the threshold of "not less than 60%" required by the specifications, the compliance judgment result of "this design scheme meets the current requirements of the sponge city specifications" is output. If the reduction rate obtained from the simulation is lower than the threshold of the specifications, the optimization direction of the design parameters is simultaneously output, including three feasible optimization paths: increasing the proportion of the rain garden catchment area, increasing the depth of the depression, and improving the saturated permeability coefficient of the planting soil layer.

[0114] like Figure 2 As shown, the present invention also provides a stormwater runoff infiltration simulation system for sponge city rain gardens. The system serves as the carrier of the above method and specifically includes five functional modules that are sequentially connected in communication. The specific functions of each module are as follows:

[0115] Data Input and Preprocessing Module: This module performs all the functions of step S1 above, including inputting centimeter-level 3D terrain models, LID facility layered structure and geotechnical parameters, minute-by-minute rainfall curves for design storms, standardization processing, and construction of integrated datasets.

[0116] Infiltration model construction module: This module performs all the functions of step S2 above, constructs an improved Green-Ampt infiltration model, and completes the real-time infiltration calculation of each grid cell throughout the entire rainfall process.

[0117] Surface runoff simulation module: This module performs all the functions of step S3 above. It constructs a two-dimensional shallow water equation surface runoff calculation model based on a centimeter-level three-dimensional terrain model, and completes the simulation and solution of the spatiotemporal collection and evolution process of surface runoff in the catchment area.

[0118] Two-way coupled simulation module: used to perform all the functions of the above S4 step, realize the two-way time coupling of the improved Green-Ampt infiltration model and the two-dimensional shallow water equation model, and complete the minute-by-minute dynamic simulation of the entire hydrological process of the design rainstorm.

[0119] The Results Output and Compliance Verification Module is used to perform all the functions of the above S5 steps, complete the output of dynamic visualization results and hydrological quantitative indicators, and conduct compliance assessment and judgment with sponge city design specifications.

[0120] This embodiment accurately reproduces the dynamics of infiltration and overflow in a rain garden throughout the entire design storm process using the technical solution of the present invention, and accurately predicts the overflow initiation time and the temporal distribution of overflow volume. To intuitively demonstrate the core advantages of the present invention compared with the prior art, the simulation results of the method of the present invention are compared with those of the traditional runoff coefficient method commonly used in engineering. The comparison data are detailed in Table 2.

[0121] Table 2 Comparison of simulation results between the method of the present invention and the traditional runoff coefficient method.

[0122] Contrast index The method of the present application Traditional runoff coefficient method Specification requirement threshold value Runoff peak flow (m³ / s) 0.027 0.022 - Runoff peak reduction rate 62.5% 69.4% ≥60% Overflow starting time (minutes after rainfall) 14 Unable to predict - Maximum water storage depth of rainwater garden (mm) 280 300 (generalized value) - Relative error with measured value ≤8% ≥42% -

[0123] As shown in Table 2, the traditional runoff coefficient method completely ignores the dynamic process of overflow when LID facilities are full, and can only provide a static estimate of the peak reduction rate. Moreover, the estimated result is significantly overestimated, which can easily lead to designers misjudging the compliance of the scheme. At the same time, it cannot predict key engineering parameters such as the overflow initiation time. In contrast, the method of this invention can accurately restore the real hydrological dynamics of the entire rainstorm process, greatly reducing the calculation error. At the same time, it can output key engineering data such as overflow sequence and water storage time, forming a compliance verification closed loop that directly matches the standards. This solves the core pain points of traditional experience-based design methods, which cannot predict overflow risks and cannot verify the matching with the drainage capacity of the pipe network. It fully meets the engineering needs of refined design of LID facilities in sponge cities.

[0124] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for simulating stormwater runoff infiltration in a sponge city rain garden, characterized in that, Includes the following steps: S1. Obtain a centimeter-resolution 3D terrain model of the target catchment area, hierarchical structure data of LID facilities, geotechnical engineering parameters, minute-by-minute rainfall intensity-duration curves of the design storm, and construct an integrated simulation base dataset. S2. Based on the layered structure of LID facilities, an improved Green-Ampt infiltration model is constructed to calculate the real-time infiltration of LID facilities and surrounding underlying surfaces during rainfall. S3. Based on high-precision micro-topography elevation data of centimeter-level three-dimensional terrain model, construct a two-dimensional shallow water equation-driven surface runoff calculation model to simulate the spatiotemporal convergence and evolution process of surface runoff in the catchment area. S4. The improved Green-Ampt infiltration model and the two-dimensional shallow water equation model are integrated through a two-way time-series coupling mechanism to perform minute-by-minute dynamic simulation of the entire process of the design storm, covering the entire hydrological process of surface runoff collection, water storage and infiltration of LID facilities and overflow. S5. Based on the coupled simulation results, output the dynamic visualization results of the spatiotemporal distribution of runoff throughout the entire rainstorm process, the hydrological quantitative indicators of LID facilities, and the compliance assessment results with the sponge city drainage design specifications.

2. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 1, characterized in that: Step S1 specifically includes: S11. Obtain a three-dimensional terrain model of the target catchment area with centimeter-level resolution through three-dimensional terrain scanning. The model includes micro-topographic elevation data of the entire catchment area, spatial boundaries of LID facilities, elevation data of each position of the concave edge, spatial location and elevation data of drainage outlets, and canopy interception parameters corresponding to plant configuration. S12. Collect layered structural data of LID facilities, including the layer thickness data of planting soil layer, gravel cushion layer and drainage layer; collect corresponding geotechnical engineering parameters, including saturated permeability coefficient, field water holding capacity, saturated water content and wetting front matrix suction head of each layer medium. S13. Obtain the minute-by-minute rainfall intensity-duration curve of the design storm for the corresponding design return period, and complete the construction of the integrated simulation basic dataset.

3. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 1, characterized in that: Step S2, the construction of the improved Green-Ampt infiltration model and the calculation of real-time infiltration volume, specifically includes: For the layered media structure of LID facilities, an improved Green-Ampt infiltration model is constructed, considering surface water depth-driven infiltration, stratified wetting front propagation, and interlayer percolation effect. The infiltration rate f(t) at any calculation time t is calculated using the following formula: ； In the formula: K s h is the saturated permeability coefficient of the infiltration layer where the current wetting front is located. w To calculate the real-time water depth within the LID facility at any given time; h s ψ is the thickness of the saturated soil layer above the infiltration layer where the current wetting front is located; f The matrix suction head at the wetting front of the current infiltration medium; z f (t) represents the propagation depth of the wetting front within the current infiltration layer at time t; when the wetting front penetrates the current infiltration layer and enters the next layer, the K value of the corresponding layer is updated. s With ψ f The calculation parameters are adjusted to account for the interlayer percolation effect caused by the difference in permeability coefficients between upper and lower layers, and the infiltration rate calculation results are corrected. For the surrounding underlying surface outside the LID facility, the real-time infiltration rate is calculated using a single-layer Green-Ampt model with corresponding underlying surface medium parameters. Based on the infiltration rate at each time step, the cumulative infiltration rate for the corresponding time period is obtained by integration.

4. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 1, characterized in that: Step S3 involves the construction and solution of the two-dimensional shallow water equation-driven surface runoff calculation model, specifically including: Based on a 3D terrain model with centimeter-level resolution, the target catchment area is meshed. The bottom elevation of each mesh cell is directly taken from the corresponding location micro-topographic elevation data of the 3D terrain model. The conservation form of the two-dimensional shallow water equations is then constructed as follows: ； In the formula: U is the vector of conserved variables. , where h is the real-time water depth within the grid cell, and u and v are the cross-sectional average velocity components in the x and y directions, respectively; F(U) is the flux vector in the x-direction. , where g is the acceleration due to gravity; G(U) is the flux vector in the y-direction. ; S(U) is the source term vector, which includes the bottom slope source term, friction source term, rainfall source term, and infiltration sink term.

5. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 1, characterized in that: The bidirectional time-series coupling mechanism in step S4 employs a unified minute-by-minute calculation time step that matches the design rainstorm time series, and executes the following bidirectional coupling calculation loop within each time step: The first step is to calculate the real-time infiltration and cumulative infiltration of each grid cell in the current time step based on the rainfall intensity at the current time step and the global surface water depth distribution obtained from the previous time step, using the improved Green-Ampt infiltration model. The cumulative infiltration is then used as the infiltration sink term in the governing equation of the two-dimensional shallow water equation. The second step is to substitute the updated infiltration sink term into the two-dimensional shallow water equation governing equation to solve for the water depth distribution, flow velocity distribution, and real-time water storage depth within the LID facility range of each grid cell in the current time step. The third step is to feed back the real-time water depth within the LID facility to the improved Green-Ampt infiltration model, updating the h value used to calculate the infiltration rate for the next time step. w Parameters are used to complete a single bidirectional coupling calculation loop; Repeat the above bidirectional coupled calculation loop until the minute-by-minute dynamic simulation of the entire design rainstorm process is completed.

6. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 5, characterized in that: In the second step of the bidirectional coupled calculation loop, the overflow process simulation of the LID facility is performed synchronously, specifically including: Based on a centimeter-level 3D terrain model, the overflow elevation threshold of each grid cell at the recessed edge of the LID facility is pre-obtained; After obtaining the water depth distribution at each time step, the real-time water depth of each grid cell at the edge of the LID facility is compared with the overflow elevation threshold at the corresponding location. When the real-time water depth exceeds the overflow elevation threshold, it is determined that an overflow has occurred at that location. The overflow flow rate of the grid cell is calculated and incorporated into the flux calculation of the two-dimensional shallow water equation. The overflow start time, the overflow flow rate process line per minute, and the cumulative overflow flow rate are output synchronously.

7. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 1, characterized in that: The results of the dynamic visualization of runoff spatiotemporal distribution in step S5 specifically include: The animation of water depth distribution throughout the entire rainstorm process, built on a centimeter-level 3D terrain model, uses continuous color levels to represent the real-time water depth values ​​of different grid units. Simultaneously, runoff direction vectors, overflow locations, and overflow volume indicators are overlaid to fully present the spatiotemporal evolution of the entire process of surface runoff collection, LID facility water storage, and overflow.

8. The method for simulating stormwater runoff infiltration in a sponge city rain garden according to claim 7, characterized in that: Step S5, the hydrological quantitative indicators and compliance assessment, specifically includes: The hydrological quantitative indicators include the time history curves of the water storage depth, runoff infiltration, overflow, total runoff and peak runoff at the outlet of the catchment area of ​​the LID facility. The compliance assessment compares the peak runoff reduction rate calculated by simulation with the threshold corresponding to the sponge city design code, outputs the judgment result of whether the current LID facility design scheme meets the code requirements, and outputs the direction of design parameter optimization when the code requirements are not met.

9. A dynamic simulation system for stormwater runoff infiltration in a sponge city rain garden, applied to the dynamic simulation method for stormwater runoff infiltration in a sponge city rain garden as described in any one of claims 1-8, characterized in that, Includes the following modules that are connected in sequence: The data input and preprocessing module is used to acquire a centimeter-resolution three-dimensional terrain model of the target catchment area, layered structural data of LID facilities, geotechnical engineering parameters, and minute-by-minute rainfall intensity-duration curves of the design storm, and to build an integrated simulation base dataset. The infiltration model construction module is used to build an improved Green-Ampt infiltration model based on the layered structure of LID facilities, and to calculate the real-time infiltration of LID facilities and surrounding underlying surfaces during rainfall. The surface runoff simulation module is used to construct a two-dimensional shallow water equation-driven surface runoff calculation model using high-precision micro-topographic elevation data from a centimeter-level three-dimensional terrain model, simulating the spatiotemporal convergence and evolution of surface runoff within the catchment area. The two-way coupled simulation module is used to integrate the improved Green-Ampt infiltration model with the two-dimensional shallow water equation model through a two-way time-series coupling mechanism to perform minute-by-minute dynamic simulation of the entire process of the design storm, covering the entire hydrological process of surface runoff collection, water storage and infiltration of LID facilities and overflow. The results output and compliance verification module is used to output dynamic visualization results of runoff spatiotemporal distribution throughout the entire rainstorm process, hydrological quantitative indicators of LID facilities, and compliance assessment results with sponge city drainage design specifications, based on the coupled simulation results.

Images