Basin water environment simulation method and device
By dividing the grid according to the water depth threshold in the water basin water environment simulation and realizing dynamic bidirectional coupling between the land model and the water body model, the problem of unclear separation response process and coupling boundaries of the single model is solved, and a more scientific water basin water environment simulation is achieved. Accurate description of the interaction between pollutants.
Patent Information
- Application Number
- CN202210103081.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-27
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing single land model or water body model cuts the basin water-land hydrology and water quality response process when simulating water environment pollution in the basin, and there are problems of unclear coupling boundaries and unconserved fluxes in the coupling between the land model and the water body model.
By meshing the basin according to the preset water depth threshold, the deeper water depth is used as the submerged area, the water body model is used to simulate the convection diffusion process of pollutants, the shallower water depth is used as the non-submerged area, and the land model is used to simulate pollution. The transfer process of objects realizes direct dynamic bidirectional coupling between the land model and the water body model, and dynamically updates the coupling boundary to ensure the dynamic changes of pollutants and water flow.
A more scientific water environment simulation of the basin is achieved, and the boundary conditions can be automatically adjusted to ensure the mutual influence of pollutants and water flow, the scientificity and accuracy of the simulation are improved, and the boundary problem in the one-way coupling method is overcome.
Smart Images

Figure CN114707293B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of water environment simulation, and more specifically, to a method and device for simulating a watershed water environment. Background Art
[0002] With the process of urbanization and rapid economic development, water pollution in my country is a prominent problem, which directly threatens people's health and life safety. The comprehensive management of river basin water environment has been highly valued by management departments at all levels. River basin water environment simulation will provide a scientific basis for the design and implementation of water environment management programs.
[0003] As an important means of analyzing and simulating water environment in a river basin, water environment simulation based on numerical models and its application have become a hot topic in related research. According to the different functional uses of water environment models, water environment models can be divided into land models, water body models, and coupled models of land models and water body models. The land model only calculates the accumulation and washoff of non-point source pollution, and obtains the average value of the land pollution load in time and space (EMC, Event Mean Concentration), but lacks the movement process of the land pollution load with rainfall convergence, that is, it lacks a land pollutant transport module. The widely used land models include SWMM, SWAT, HSPF, etc., but these models lack the land pollution load with hydrological convergence The module cannot predict the detailed spatiotemporal variation information of the land pollution load and can only be unidirectionally coupled with the water body model; the water body model includes hydrodynamic models and water quality models, which are mainly used to simulate the convection and diffusion process of pollutants in rivers, lakes, reservoirs, estuaries, coastal waters, etc., and obtain detailed distribution information of pollutants in water bodies. The widely used water body models include WASP, EFDC, MIKE series, etc., but these models lack the mechanism analysis of the land pollution load entering the water body and the feedback mechanism between the water body process and the land process. Based on a single land model or water body model, the hydrological elements, nutrient load and convection and diffusion in the water body at the basin outlet can be simulated, which separates the water-land hydrology and water quality response processes in the basin.
[0004] Compared with a single land or water model, the dynamic bidirectional coupling model of land and water achieves a comprehensive simulation of pollutants in land and river and lake water bodies by coupling land and water models. The coupling modes of land models and water models are mainly divided into external coupling and internal coupling. In external coupling, the land model and the water model operate independently, and the output of the land model serves as the input of the water model to drive the operation of the water model, which is a one-way coupling. This type of coupling model is easy to develop and has a fast calculation speed, which is suitable for large-scale watershed water environment simulation. However, the number of coupling boundary points of this method is limited, and the coupling boundary points do not change with the water level fluctuations of rivers, lakes, and reservoirs. The flow and pollutant load at the boundary points may be exaggerated. In internal coupling, the land model and the water model run synchronously in time. The results of the land model provide boundary conditions for the water model. The results of the water model affect the simulation of the land model through the boundary conditions, thereby realizing dynamic two-way coupling between the land model and the water model. However, since the coupling boundary of the internal coupling mode is fixed, when the grids of the land model and the water model do not match, the water level, flow, and pollutant variables at the fixed coupling boundary points need to be realized through interpolation, and the conservation of flux through the coupling boundary cannot be guaranteed. Summary of the invention
[0005] In view of the above problems, the purpose of the present invention is to provide a method and device for simulating the water environment of a river basin, so as to overcome the deficiency of using a single land model or water body model to simulate the water environment pollution of the river basin, which separates the water-land hydrological and water quality response process of the river basin, and overcome the problems of unclear coupling boundary and non-conservation of coupling interface flux in the current coupling between land model and water body model.
[0006] The present invention provides a method for simulating a watershed water environment, comprising the following steps:
[0007] S1. According to a preset water depth threshold, the watershed to be predicted is divided into grids, and the grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the grids with water depths greater than the preset water depth threshold are regarded as flooded grids;
[0008] S2, connecting the land model and the water body model by using the boundary between the non-flooded area formed by the non-flooded grid and the flooded area formed by the flooded grid as a coupled moving boundary;
[0009] S3. For the non-flooded grid, the land model is used to simulate the generation of pollutants on land and the transport process with runoff, and the pollutant load distribution data of the non-flooded area is calculated; for the flooded grid, the pollutant load distribution data of the non-flooded area is used as a calculation condition, and the water body model is used to simulate the convection diffusion process of pollutants in the water body, and the concentration distribution data of pollutants in the flooded area is calculated;
[0010] S4. Calculate the water depth data of the submerged grid and the non-submerged grid based on the time step of model calculation, and update the coupled moving boundary according to the water depth data, and return to step S2.
[0011] In addition, a preferred solution is to divide the watershed to be predicted into grids according to a preset water depth threshold, and to divide the grids whose water depth is lower than the preset water depth threshold into non-flooded grids, and the grids whose water depth is greater than the preset water depth threshold into flooded grids, including:
[0012] Dividing the watershed to be predicted into uniform rectangular grids;
[0013] Defining the boundaries of the rectangular grids and the relationship between adjacent rectangular grids, and numbering the rectangular grids based on the definitions;
[0014] The variables are stored in the center position of the numbered rectangular grids, initial conditions are set for each numbered rectangular grid, and the numbered rectangular grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the numbered rectangular grids with water depths greater than the preset water depth threshold are regarded as flooded grids.
[0015] In addition, a preferred solution is that the land model includes:
[0016] A hydrological model for calculating the convergence velocity and water depth of the non-flooded grid, a non-point source pollution load model for calculating the accumulation and flushing amount of pollutants in the non-flooded grid according to the convergence velocity and water depth of the non-flooded grid, and a pollutant transport model for obtaining non-point source pollutant load distribution data by taking the accumulation and flushing amount of pollutants in the non-flooded grid as source terms;
[0017] The water body model includes:
[0018] The method is used to use the confluence velocity and water depth of the non-flooded grid as driving conditions for the movement of water flow in the water body, obtain a hydrodynamic model of the flow velocity and water depth in the flooded grid by solving the two-dimensional shallow water equation in the local area, and use the concentration distribution data of non-point source pollutants in the non-flooded grid as driving conditions for the convection diffusion process of pollutants in the water body, and solve the water quality model of the convection diffusion process of pollutants in the water body based on the flow velocity and water depth in the flooded grid.
[0019] In addition, a preferred solution is that, in the process of using the land model to simulate the generation of pollutants on land and the transport process with runoff for the non-flooded grid and calculating the pollutant load distribution data of the non-flooded area,
[0020] In the hydrological model:
[0021] The rainfall of the non-flooded grid watershed is calculated by the inverse distance interpolation method, and the infiltration of the watershed is calculated by the Green-Hampton formula;
[0022] Calculating the net rainfall of the non-flooded grid based on the rainfall and infiltration of the watershed;
[0023] The diffusion wave equation is used to calculate the convergence velocity and water depth of the non-flooded grid;
[0024] In the non-point source pollution load model, based on the convergence velocity and water depth calculated by the hydrological model, an exponential equation is used to calculate the amount of pollutants entering the water body with rainfall runoff;
[0025] In the pollutant transport model:
[0026] The water quality equation is used to simulate the transport process of pollutants on land with water flow, and the water quality equation is discretized using a finite volume algorithm of the Godunov center format to obtain the pollutant load distribution data of the non-flooded area.
[0027] In addition, a preferred solution is that, in the process of using the water body model to simulate the convection diffusion process of pollutants in the water body for the submerged grid, and using the pollutant load distribution data of the non-submerged area as a calculation condition to calculate the concentration distribution data of the pollutants in the submerged area,
[0028] The hydrodynamic model and the water quality model are coupled and solved by a tightly coupled solution method, and the variable at the center point of the submerged grid is interpolated to the interface of the submerged grid by a MUSCL linear interpolation method;
[0029] The Godunov format finite volume method is used for spatial discretization, the HLLC approximate Riemann solver is used to solve the convective flux on the submerged grid interface, the second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, and the water depth and flow velocity of the grid interface of the submerged area are solved;
[0030] Based on the conservation of mass and momentum of the coupled moving boundary interface, the pollutant load distribution of the non-flooded area is used as the calculation condition of the water body model, and the water depth and flow velocity of the flooded grid are substituted into the water quality equation as known quantities. The Godunov format is used for spatial discretization, and the HLLC approximate Riemann solution is used to solve the flux of pollutants passing through the flooded grid interface. The second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, so as to obtain the concentration distribution data of pollutants in the flooded area.
[0031] In addition, a preferred solution is that the method for determining the time step of the model calculation includes:
[0032] The time step of the model calculation is determined by using an explicit grid discretization method; wherein the time step constraint is:
[0033] in,
[0034] Where C is a constant used to ensure the stability of the format, 1 / 2≤C<1; u n i,j ,v n i,j are the velocity components of the current unit in the x and y directions (m·s -1 );h n i,j is the water depth of the unit grid (m); g is the gravity velocity; D x and D y are the diffusion coefficients in the x and y directions (m 2 ·s -1 ), γ is the stability coefficient of the diffusion term.
[0035] In addition, a preferred solution is to calculate the water depth data of the submerged grid and the non-submerged grid based on the time step of the model calculation, and update the coupled moving boundary according to the water depth data, including:
[0036] calculating characteristic waves at discontinuities between the submerged grid and the non-submerged grid, and determining the direction of mass and momentum transfer between the submerged area and the non-submerged area according to the direction of the characteristic waves;
[0037] updating the water depth and flow velocity of the non-flooded area and the flooded area according to the mass and momentum transfer direction between the flooded area and the non-flooded area;
[0038] According to the water depths of the non-flooded area and the flooded area, taking the grids whose water depths are lower than the preset water depth threshold as updated non-flooded grids, and taking the grids whose water depths are greater than the preset water depth threshold as updated flooded grids;
[0039] The coupled moving boundary connecting the land model and the water body model is updated according to the boundary between the new non-flooded area formed by the updated non-flooded grid and the new flooded area formed by the updated flooded grid.
[0040] In addition, a preferred solution is that the calculating of the characteristic waves at the discontinuities between the submerged grid and the non-submerged grid, and determining the mass and momentum transfer direction between the submerged area and the non-submerged area according to the direction of the characteristic waves comprises:
[0041] When the directions of the characteristic waves at the coupled moving boundary all point to the submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the non-submerged grid, and it is determined that the non-submerged area transfers mass to the submerged area but does not transfer momentum;
[0042] When the characteristic wave at the coupled moving boundary is distributed in both the submerged grid and the non-submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the submerged grid and the water depth and flow velocity of the non-submerged grid, and it is determined that the non-submerged area transfers mass to the submerged area and the submerged area transfers momentum to the non-submerged area;
[0043] When the directions of the characteristic waves at the coupled moving boundary all point to the non-submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the submerged grid, and it is determined whether the submerged area transfers mass and momentum to the non-submerged area.
[0044] The basin water environment simulation device provided by the present invention comprises:
[0045] A grid division module is used to divide the watershed to be predicted into grids according to a preset water depth threshold, and to divide the grids whose water depth is lower than the preset water depth threshold into non-flooded grids, and the grids whose water depth is greater than the preset water depth threshold into flooded grids;
[0046] A model building module, used to build a land model and a water body model by taking the boundary between the non-flooded area formed by the non-flooded grid and the flooded area formed by the flooded grid as a coupled dynamic boundary;
[0047] A calculation module, for simulating the generation of pollutants on land and the transport process of pollutants with runoff using the land model for the non-flooded grid, and calculating the pollutant load distribution data of the non-flooded area; for the flooded grid, using the pollutant load distribution data of the non-flooded area as a calculation condition, using the water body model to simulate the convection diffusion process of pollutants in the water body, and calculating the concentration distribution data of pollutants in the flooded area;
[0048] The grid updating module is used to calculate the water depth data of the submerged grid and the non-submerged grid based on the time step of the model calculation, and update the coupled moving boundary according to the water depth data.
[0049] It can be seen from the above technical solutions that the basin water environment simulation method and device provided by the present invention overcomes the shortcomings of using a single land model or water model and a one-way coupling of a land model and a water model to simulate basin water environment pollution by directly and dynamically bidirectionally coupling a land model with a water model. By judging the state of the grid according to a preset water depth threshold, the area with a deeper water depth is regarded as a flooded area, and the convection diffusion process of pollutants occurs. The water body model is used to calculate the water depth and flow rate of the flooded area, and the concentration of pollutants is solved. The area with a shallower water depth is regarded as a non-flooded area, and the transfer process of pollutants occurs. The land model is used to simulate the generation and transfer process of pollutants on land. The dynamic bidirectional coupling of the two models established does not specify the boundary position in the calculation. The location where pollutants and water flow enter the flooded area is dynamically changing. Flood propagation can also flood the original confluence point. The boundary shift forms a new coupling boundary, realizing the alternating conversion of the two areas, and the simulation is closer to the real situation. Compared with the external single coupling method, which requires the calculation of the land model and the water model in sequence, the two models are calculated simultaneously in the bidirectional coupling method adopted by the present invention. Pollutants can flow from the non-flooded area into the flooded area, and pollutants in rivers and lakes can also invade the land area of water collection, thereby realizing the mutual influence of pollutants on the shore and in the water body, which is more in line with physical reality and more scientific. The coupled moving boundary is used as the boundary of the control volume unit, and different calculation models (land model and water body model) are used on both sides of the moving boundary. Through the finite volume format of water volume, momentum, and pollutant mass conservation, it not only ensures the conservation of various physical quantities, but also can automatically realize the land pollutants entering the water body through the coupled moving boundary, realizing the driving of the water model by the land model, instead of setting boundary conditions artificially and loading land pollutants as boundary conditions into the water model.
[0050] In order to achieve the above and related purposes, one or more aspects of the present invention include the features that will be described in detail later. The following description and the accompanying drawings describe some exemplary aspects of the present invention in detail. However, these aspects indicate only some of the various ways in which the principles of the present invention can be used. In addition, the present invention is intended to include all these aspects and their equivalents. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] By referring to the following description in conjunction with the accompanying drawings, and with a more comprehensive understanding of the present invention, other objects and results of the present invention will become more apparent and easy to understand. In the accompanying drawings:
[0052] Figure 1 A schematic diagram of a flow chart of a method for simulating a watershed water environment according to an embodiment of the present invention;
[0053] Figure 2 A schematic diagram of a moving process of a coupled moving boundary interface according to an embodiment of the present invention;
[0054] Figure 3 A flow chart of basin water environment model calculation according to an embodiment of the present invention;
[0055] Figure 4 A schematic diagram of defining the relationship between grid boundaries and adjacent grids according to an embodiment of the present invention;
[0056] Figure 5 A schematic diagram of a land pollutant load calculation process according to an embodiment of the present invention;
[0057] Figure 6 A schematic diagram of a MUSCL linear interpolation format according to an embodiment of the present invention;
[0058] Figure 7 Schematic diagram of characteristic lines on the left and right sides of the x-direction grid interface according to an embodiment of the present invention;
[0059] Figure 8 Schematic diagram of characteristic lines on the left and right sides of the interface in the y direction according to an embodiment of the present invention;
[0060] Fig. 9 A two-dimensional slope watershed topographic distribution map according to an embodiment of the present invention;
[0061] Fig.10 A two-dimensional slope watershed grid discrete graph according to an embodiment of the present invention;
[0062] Fig.11 A schematic diagram of pollutant distribution at the initial moment according to an embodiment of the present invention;
[0063] Fig.12 A schematic diagram of a flow process at a basin outlet simulated by different models according to an embodiment of the present invention;
[0064] Fig.13 Schematic diagram of pollutant concentration distribution at different times according to an embodiment of the present invention (at=4200s; bt=6300s);
[0065] Fig.14 Schematic diagram of pollutant mass conservation statistics according to an embodiment of the present invention.
[0066] The same reference numerals in the drawings indicate similar or corresponding features or functions. DETAILED DESCRIPTION
[0067] In the following description, for the purpose of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more embodiments. However, it is apparent that these embodiments may also be implemented without these specific details.
[0068] In response to the aforementioned problems that the current water environment simulation technology uses a one-way coupling method of land models and water models, which ignores the impact of water body processes on land processes, separates the water-land hydrology and water quality response processes in the basin, is not scientific enough, and has low calculation accuracy, a dynamic two-way coupling watershed water environment simulation method and device are proposed.
[0069] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0070] In order to illustrate the basin water environment simulation method and device provided by the present invention, Figure 1 The flow chart of the method for simulating water environment in a watershed according to an embodiment of the present invention is shown; Figure 2 The moving process of the coupled moving boundary interface according to an embodiment of the present invention is shown; Figure 3 The calculation process of the watershed water environment model according to the embodiment of the present invention is shown; Figure 4 The definition of the relationship between the grid boundary and adjacent grids according to an embodiment of the present invention is shown; Figure 5 The land pollutant load calculation process according to an embodiment of the present invention is shown; Figure 6 The MUSCL linear interpolation format according to an embodiment of the present invention is shown; Figure 7 The characteristic lines on the left and right sides of the x-direction grid interface according to an embodiment of the present invention are shown; Figure 8 The characteristic lines on the left and right sides of the y-direction interface according to an embodiment of the present invention are shown; Fig. 9 The two-dimensional slope watershed topographic distribution according to an embodiment of the present invention is shown; Fig.10 The two-dimensional slope watershed grid discretization according to an embodiment of the present invention is shown; Fig.11 shows the pollutant distribution at the initial moment according to an embodiment of the present invention; Fig.12 The flow process of the basin outlet simulated by different models according to the embodiments of the present invention is shown; Fig.13 The concentration distribution of pollutants at different times according to an embodiment of the present invention is shown (at=4200s; bt=6300s); Fig.14 Pollutant mass conservation statistics according to an embodiment of the present invention are shown.
[0071] like Figure 1 As shown, the basin water environment simulation method provided by the present invention comprises the following steps:
[0072] S1. According to a preset water depth threshold, the watershed to be predicted is divided into grids, and the grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the grids with water depths greater than the preset water depth threshold are regarded as flooded grids;
[0073] S2, connecting the land model and the water body model using the boundary between the non-flooded area formed by the non-flooded grid and the flooded area formed by the flooded grid as a coupled moving boundary;
[0074] S3. For non-flooded grids, the land model is used to simulate the generation of pollutants on land and the transport process with runoff, and the pollutant load distribution data of the non-flooded area is calculated; for flooded grids, the pollutant load distribution data of the non-flooded area is used as the calculation condition, and the water body model is used to simulate the convection and diffusion process of pollutants in the water body, and the concentration distribution data of pollutants in the flooded area is calculated;
[0075] S4. Based on the time step of model calculation, calculate the water depth data of the submerged grid and the non-submerged grid, and update the coupled moving boundary according to the water depth data, and return to step S2.
[0076] Among them, in the process of establishing the land model and the water body model in step S2, it is preferred to establish the land model in the non-flooded area and the water body model in the flooded area, so that the land model can be used in the non-flooded grid to simulate the generation of pollutants on land and their transport with runoff; and the water body model can be used in the flooded grid to simulate the convection diffusion process of pollutants in the water body.
[0077] By directly and dynamically bidirectionally coupling the land model with the water model, the shortcomings of using a single land model or water model, as well as the one-way coupling of the land model and the water model, in predicting and simulating water pollution in the basin are overcome. By judging the state of the grid according to the preset water depth threshold, the area with deeper water depth is regarded as the flooded area, and the convection and diffusion process of pollutants occurs. The water body model is used to calculate the water depth and flow rate of the flooded area, and the concentration of pollutants is solved. The area with shallower water depth is regarded as the non-flooded area, and the transfer process of pollutants occurs. The land model is used to simulate the generation and transfer process of pollutants on land. The dynamic bidirectional coupling of the two models is established. The boundary position is not specified in the calculation. The location where pollutants and water flow enter the flooded area is dynamically changing. Flood propagation can also flood the original confluence point. The boundary shift forms a new coupling boundary, realizing the alternating conversion of the two areas, and the simulation is closer to the real situation. Compared with the external single coupling method, which requires the calculation of the land model and the water model in sequence, the two models are calculated simultaneously in the bidirectional coupling method adopted by the present invention. Pollutants can flow from the non-flooded area into the flooded area, and pollutants in rivers and lakes can also invade the land area of water collection, thereby realizing the mutual influence of pollutants on the shore and in the water body, which is more in line with physical reality and more scientific. The coupled moving boundary is used as the boundary of the control volume unit, and different calculation models (land model and water body model) are used on both sides of the moving boundary. Through the finite volume format of water volume, momentum, and pollutant mass conservation, it not only ensures the conservation of various physical quantities, but also can automatically realize the land pollutants entering the water body through the coupled moving boundary, realizing the driving of the water model by the land model, instead of setting boundary conditions artificially and loading land pollutants as boundary conditions into the water model.
[0078] As a preferred solution of the present invention, according to a preset water depth threshold, the watershed to be predicted is divided into grids, grids with water depths lower than the preset water depth threshold are used as non-flooded grids, and grids with water depths greater than the preset water depth threshold are used as flooded grids, including:
[0079] Divide the watershed to be predicted into uniform rectangular grids;
[0080] The boundaries of the rectangular grid and the relationship between adjacent rectangular grids are defined, and the rectangular grids are numbered based on the definition;
[0081] The variables are stored in the center of the numbered rectangular grids, initial conditions are set for each numbered rectangular grid, and the numbered rectangular grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the numbered rectangular grids with water depths greater than the preset water depth threshold are regarded as flooded grids.
[0082] The uniform rectangular grid is used. Since the structured grid has the advantages of easy implementation, fast generation speed, good grid quality and simple data structure, it can be well matched with the terrain grid data. The model can directly read the data information in GIS format, such as DEM, land use information, soil data, etc., to improve the efficiency of data pre-processing. Therefore, the preferred embodiment of the present invention uses a structured uniform rectangular grid to discretize the calculation area. First, the calculation area is divided into a uniform rectangular grid, and a preset water depth threshold is set. The calculation domain is divided into a flooded area and a non-flooded area according to the grid water depth. When the water depth of the grid is greater than the preset water depth threshold, the area is regarded as a flooded area; if the water depth of the grid is less than the threshold, the area is regarded as a non-flooded area. The range of the flooded area and the non-flooded area changes with time, and the coupling boundary between the two is also in dynamic change. Different models are used in different areas. In the non-flooded area, the land model is used to simulate the generation and transport process of pollutants; in the flooded area, the water body model is used to simulate the convection diffusion process of pollutants in the water.
[0083] As a preferred embodiment of the present invention, the initial conditions at least include elevation, water depth, slope, roughness, external pollution load and pollutant diffusion coefficient, and of course, other conditions may be included according to actual needs.
[0084] As a preferred embodiment of the present invention, the land model includes:
[0085] A hydrological model for calculating the convergence velocity and water depth of non-flooded grids, a non-point source pollution load model for calculating the accumulation and washoff of pollutants in non-flooded grids based on the convergence velocity and water depth of non-flooded grids, and a pollutant transport model for obtaining non-point source pollutant load distribution data by taking the accumulation and washoff of pollutants in non-flooded grids as source terms;
[0086] The water model includes:
[0087] It is used to take the confluence velocity and water depth of the non-flooded grid as the driving conditions for the water flow movement in the water body, and obtain the hydrodynamic model of the flow velocity and water depth in the flooded grid by solving the two-dimensional shallow water equation in the local area. It is used to take the concentration distribution data of non-point source pollutants in the non-flooded grid as the driving conditions for the convection and diffusion process of pollutants in the water body, and solve the water quality model of the convection and diffusion process of pollutants in the water body according to the flow velocity and water depth in the flooded grid.
[0088] Hydrological model: The hydrological model includes runoff generation and confluence processes. The runoff generation process includes rainfall, evaporation and infiltration. In the real-time calculation, evaporation can be ignored. The rainfall is calculated by inverse distance interpolation method, and the infiltration is calculated based on the Green-Ampt formula. The net rainfall calculated in the runoff generation process is used as the source term of the mass conservation equation. The diffusion wave equation is used for confluence calculation. This model has good numerical stability and can improve calculation efficiency. Among them, the mass conservation equation is:
[0089]
[0090] The diffusion wave equation is:
[0091]
[0092]
[0093] Where h is the water depth in meters; q x and q y are the grid single width flow in the x and y directions, in m 2 ·s -1 ;q r The runoff is the amount of rainfall minus the infiltration, that is, the net rainfall, in m·s -1 ;Q x and Q y The flow rates in the x and y directions, respectively, in m 3 ·s -1 ; A is the area of the flow cross section, in m 2 ; R is the hydraulic radius, unit is m; S is the water level slope; n is the surface roughness.
[0094] Non-point source pollution load model: The pollution load model includes the accumulation and washout process of pollutants. During the dry and rainless period, pollutants accumulate and grow on the land. An exponential function is used to describe the accumulation process of pollutants on the land during drought. It is believed that the accumulation of pollutants follows exponential growth until the maximum value. Pollutants accumulated on the land will be washed into the runoff by rainwater during the rainy season, and then enter the canals and rivers until the accumulated amount of pollutants on the land is exhausted. The exponential washout function is used to calculate the washout amount of pollutants. The washout load is proportional to the product of the net flow to the power of n and the accumulated amount of pollutants. The pollutant accumulation equation formula is expressed as follows:
[0095]
[0096] In the formula, B k is the cumulative amount of the kth pollutant per unit area, in kg·ha -1 ; C1 is the maximum cumulative amount, unit: kg·ha -1 , C2 is the accumulation rate constant, unit day-1 The selection of parameters C1 and C2 can be determined by reference to literature or through experiments. The values of the parameters vary with the location of the study area and the composition of the pollutants. The reference value range of C1 is 1.28 to 12.44 kg·ha -1 ; The reference value range of C2 is 0.1~0.4day -1 ; t is the accumulation time of pollutants, unit is day. The accumulation equation shows that the growth of pollutants is exponentially related to time. When the growth reaches the maximum, the growth stops.
[0097] Conventional non-point source pollution load calculation will be integrated over the area of the sub-basin (or sub-region), as shown in Formula 1, to obtain the total amount of pollutants in the region W over time, as shown in Formula 2, where Formula 1 is: Formula 2 is:
[0098] In the formula, (B k ) Total is the total accumulation of the kth pollutant in the watershed, in kg; Ω is the area of the sub-region, in ha; W is the scour load, in kg / h; C3 is the scour coefficient, in mm·h -1 ) -C4 ·h -1 ; C4 is the scour index, dimensionless; q is the runoff rate per unit area, unit: mm·h -1 , The unit is h -1 .
[0099] Adopt (B k ) Total The land non-point source pollution load calculation method cannot reflect the spatial and temporal changes of pollutants in the sub-area. In order to reflect the changes of non-point source pollution load with spatial position and time, the present invention calculates the pollution load per unit area, as shown in Formula 3, where Formula 3 is:
[0100]
[0101] In the formula, S NPS,k It is the load per unit area of the kth non-point source pollutant that enters the land through rainfall and is transported with runoff, in kg·m -2 ·s; the units and meanings of the other symbols are the same as above. The selection of parameters C3 and C4 can be determined by reference to literature or through experiments. The values of the parameters vary with the location of the study area and the composition of pollutants. The reference value range of C3 is 2 to 9, and the reference value range of C4 is 0.2 to 0.6.
[0102] Pollutant transport model on land: The water quality equation is used to simulate the transport process of pollutants on land with runoff. The calculation formula is shown in Formula 4. Among them, the pollutant load per unit area calculated by Formula 3 is used as the source term of the water quality equation. Among them, Formula 4 is:
[0103]
[0104] In the formula, c k is the concentration of the kth pollutant, in kg·m -3 , h is the water depth, unit is m; u, v are the flow velocities in the x and y directions, unit is m·s -1 , calculated by the hydrological model; D x,k and D y,k Diffusion coefficient of the kth pollutant in the x and y directions, respectively, in m 2 ·s -1 ; S NPS,k is the non-point source load of the kth pollutant, in kg·m -2 ·s -1 ; S c,k The amount of the kth pollutant component generated by self-degradation or chemical reaction, in kg·m -3 ·s -1 , which can generally be expressed as S c.k =-λhc, λ is the self-purification coefficient of the water body.
[0105] Formula 4 makes it possible to calculate the spatial distribution of land pollutants by adding land non-point source terms. When the land model is coupled with the water model, the pollutant concentration information obtained after applying Formula 4 on land to calculate the transport process of land pollutants will be used as the condition for calculating the convection and diffusion of pollutants in the water body. Such conditions will more accurately describe the actual spatial location of land pollutants entering the water body, laying the foundation for realizing the dynamic two-way coupling of the land model and the water model.
[0106] Hydrodynamic model: The hydrodynamic model is calculated using the two-dimensional shallow water equation (SWE), which consists of the continuity equation and the momentum equation. The water volume change obtained by the flow calculation is used as the source term of the continuity equation. The momentum equation is used to calculate the movement of the water flow. The time step is limited by the Courant stability condition. The formula of SWE is expressed as follows:
[0107]
[0108] In the formula, U is the conserved physical quantity, F and G are the convection terms in the x and y directions respectively, and the expression is as follows:
[0109] U=[h,hu,hv] T ,F=[hu,hu 2 +gh2 / 2,huv] T G=[hv,huv,hv 2 +gh 2 / 2] T
[0110] S x and S y are the bottom slope source term and friction source term in the x and y directions respectively, expressed as follows:
[0111]
[0112] Where h is the water depth in meters; u and v are the flow velocities in the x and y directions, respectively, in meters per second. -1 ; z is the elevation, unit is m; C is the Scheherazade coefficient that represents the surface roughness, calculated by the Manning formula.
[0113] Water quality model in water body: The water quality equation is used to simulate the convection and diffusion process of pollutants in water body. The water depth and flow velocity calculated by the hydrodynamic model are substituted into the water quality equation as known quantities to obtain the concentration of pollutants. Among them, the flushing amount of land non-point source pollutants enters the water body through the coupled moving boundary, and the degradation of the pollutants themselves and the flux of external input pollutants are used as the source terms of the equation. The water quality equation ignoring the sediment release source term is used as Formula 5, where Formula 5 is as follows:
[0114]
[0115] In the formula, the meaning of symbols is the same as that of formula 4. Compared with formula 4, formula 5 ignores the pollutant release from the sediment, that is, it does not consider the non-point source term S NPS,k .
[0116] Since Formula 4 and Formula 5 have the same form, the dynamic exchange of pollutants between the land area and the water area is carried out through the internal connection boundary of the land model and the water model, ensuring the conservation of the flux of water and pollutant components on the coupled dynamic boundary.
[0117] As a preferred solution of the present invention, in the process of using a land model to simulate the generation of pollutants on land and the transport process with runoff for non-flooded grids, and calculating the pollutant concentration distribution data in the non-flooded area,
[0118] In the hydrological model:
[0119] The rainfall in the non-flooded grid watershed was calculated by the inverse distance interpolation method, and the infiltration in the watershed was calculated by the Green-Hampton formula;
[0120] Calculate the net rainfall for the non-flooded grid based on the rainfall and infiltration in the watershed;
[0121] The diffusion wave equation is used to calculate the convergence velocity and water depth of the non-flooded grid;
[0122] In the non-point source pollution load model, based on the runoff velocity and water depth calculated by the hydrological model, an exponential equation is used to calculate the amount of pollutants entering the water body with rainfall runoff;
[0123] In the pollutant transport model:
[0124] The water quality equation is used to simulate the transport of pollutants on land with water flow. The finite volume algorithm of Godunov center format is used to discretize the water quality equation to obtain the pollutant load distribution data in the non-flooded area.
[0125] As a preferred embodiment of the present invention, in the process of using a land model to simulate the generation of pollutants on land and the transport process with runoff for non-flooded grids, and calculating the pollutant load distribution data of the non-flooded area,
[0126] In the hydrological model:
[0127] The rainfall in the non-flooded grid watershed was calculated by the inverse distance interpolation method, and the infiltration in the watershed was calculated by the Green-Hampton formula;
[0128] Calculate the net rainfall for the non-flooded grid based on the rainfall and infiltration in the watershed;
[0129] The diffusion wave equation is used to calculate the convergence velocity and water depth of the non-flooded grid;
[0130] In the non-point source pollution load model, based on the runoff velocity and water depth calculated by the hydrological model, an exponential equation is used to calculate the amount of pollutants entering the water body with rainfall runoff;
[0131] In the pollutant transport model:
[0132] The water quality equation is used to simulate the transport of pollutants on land with water flow. The finite volume algorithm of Godunov center format is used to discretize the water quality equation to obtain the pollutant load distribution data in the non-flooded area.
[0133] As a preferred solution of the present invention, in the process of using a water body model to simulate the convection diffusion process of pollutants in the water body for the submerged grid, and using the pollutant load distribution data of the non-submerged area as the calculation condition to calculate the concentration distribution data of the pollutants in the submerged area,
[0134] The hydrodynamic model and the water quality model are coupled and solved by a tightly coupled solution method, and the variables at the center point of the submerged grid are interpolated to the interface of the submerged grid by using the MUSCL linear interpolation method;
[0135] The Godunov format finite volume method is used for spatial discretization, and the HLLC approximate Riemann solver is used to solve the convection flux on the submerged grid interface. The second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space. The water depth and flow velocity of the grid interface in the submerged area are solved.
[0136] Based on the conservation of mass and momentum of the coupled moving boundary interface, the pollutant load distribution in the non-flooded area is taken as the calculation condition of the water body model, and the water depth and flow velocity of the flooded grid interface are substituted into the water quality equation as known quantities. The Godunov format is used for spatial discretization, and the HLLC approximate Riemann solution is used to solve the flux of pollutants through the flooded grid interface. The second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, so as to obtain the concentration distribution data of pollutants in the flooded area.
[0137] As a preferred solution of the present invention, a method for determining the time step of model calculation includes:
[0138] The explicit grid discretization method is used to determine the time step of the model calculation; the time step constraint is:
[0139]
[0140] Where C is a constant used to ensure the stability of the format, 1 / 2≤C<1;u n i,j ,v n i,j The flow velocity components of the current unit in the x and y directions, respectively, in m·s -1 ;h n i,j is the water depth of the unit grid, in m; g is the gravity velocity; D x and D y are the diffusion coefficients in the x and y directions, respectively, in m 2 ·s -1 ; γ is the stability coefficient of the diffusion term.
[0141] The grid discretization method is performed in an explicit format, and the calculation time step is limited by the CFL (Courant-Friefrichs-Lewy) condition. In the calculation, when grids with different attributes are used for calculation, a unified time step is used in the calculation to ensure the synchronization of the calculation.
[0142] As a preferred embodiment of the present invention,
[0143] Based on the time step of the model calculation, the water depth data of the submerged grid and the non-submerged grid are calculated, and the coupled moving boundary is updated according to the water depth data, including:
[0144] Calculate the characteristic waves at the discontinuities between the submerged grid and the non-submerged grid, and determine the direction of mass and momentum transfer between the submerged area and the non-submerged area based on the direction of the characteristic waves;
[0145] Update the water depth and flow velocity in the non-flooded area and the flooded area according to the direction of mass and momentum transfer between the flooded area and the non-flooded area;
[0146] According to the water depths of the non-flooded area and the flooded area, the grids with water depths lower than a preset water depth threshold are used as updated non-flooded grids, and the grids with water depths greater than the preset water depth threshold are used as updated flooded grids;
[0147] The coupled moving boundary connecting the land model and the water body model is updated according to the boundary between the new non-flooded area formed by the updated non-flooded grid and the new flooded area formed by the updated flooded grid.
[0148] As a preferred solution of the present invention, calculating the characteristic wave at the discontinuity between the submerged grid and the non-submerged grid, and determining the mass and momentum transfer direction between the submerged area and the non-submerged area according to the direction of the characteristic wave includes:
[0149] When the directions of the characteristic waves at the coupled moving boundary all point to the submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the non-submerged grid, and it is determined that the non-submerged area transfers mass to the submerged area, but does not transfer momentum;
[0150] When the characteristic wave at the coupled moving boundary is distributed in both the submerged grid and the non-submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the submerged grid and the water depth and flow velocity of the non-submerged grid, and it is determined that the non-submerged area transfers mass to the submerged area and the submerged area transfers momentum to the non-submerged area;
[0151] When the directions of the characteristic waves at the coupled moving boundary all point to the non-submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the submerged grid, and the mass and momentum transferred from the submerged area to the non-submerged area is determined.
[0152] The mass transfer between the flooded area and the non-flooded area is calculated by water balance, and the momentum transfer is calculated by flow velocity.
[0153] The present invention provides a watershed water environment simulation device, comprising:
[0154] A grid division module is used to divide the watershed to be predicted into grids according to a preset water depth threshold, and to divide the grids whose water depth is lower than the preset water depth threshold into non-flooded grids, and the grids whose water depth is greater than the preset water depth threshold into flooded grids;
[0155] A model building module is used to build a land model and a water body model by taking the boundary between the non-flooded area formed by the non-flooded grid and the flooded area formed by the flooded grid as a coupled dynamic boundary;
[0156] The calculation module is used to use the land model to simulate the generation of pollutants on land and the transport process with runoff for non-flooded grids, and calculate the pollutant load distribution data of non-flooded areas; for flooded grids, the pollutant load distribution data of non-flooded areas is used as the calculation condition, and the water body model is used to simulate the convection and diffusion process of pollutants in the water body, and the concentration distribution data of pollutants in the flooded area is calculated;
[0157] The grid update module is used to calculate the water depth data of the submerged grid and the non-submerged grid based on the time step of the model calculation, and update the coupled moving boundary according to the water depth data.
[0158] In order to better illustrate the basin water environment simulation method and device provided by the present invention, the specific embodiments provided are as follows:
[0159] Example 1
[0160] like Figures 2 to 14 As shown in the figure, before simulating the water environment, the hydrological data of the watershed is measured or collected through instruments, and the terrain data of the mountainous watershed, land use type data, rainfall process data and pollutant calculation related data are collected. The specific calculation process is as follows:
[0161] 1) Grid discretization and general parameter assignment: Based on structured grids, the computational domain is divided into uniform rectangular grids. Figure 4 The relationship between the grid boundary and the adjacent grids is defined. (i, j) is the current control volume number (or unit grid number), (i-1, j), (i+1, j) are the unit grids adjacent to the current unit grid on the left and right (L, R) sides in the x direction, (i, j-1), (i, j+1) are the units above and below (UP, DN) in the y direction of the current unit. The variables are stored at the center of the unit. Set initial conditions for each grid, such as elevation, water depth, slope, roughness, external pollution load, pollutant diffusion coefficient, etc.
[0162] 2) Grid status judgment. Set a preset water depth threshold. When the water depth of the grid is less than the threshold, it is regarded as a non-flooded area, and the land model is used to solve the transport process of pollutants on land; when the water depth of the grid is greater than the threshold, it is regarded as a flooded area, and the water body model is used to simulate the convection and diffusion process of pollutants in the water body.
[0163] 3) Determine the time step Δt of the model calculation. The grid discretization method uses an explicit format, and the calculation time step is limited by the CFL (Courant-Friefrichs-Lewy) condition. In the calculation, when using grids with different attributes, in order to ensure the synchronization of the calculation, a unified time step Δt is used in the calculation. The restrictions are as follows:
[0164]
[0165] Where C is a constant used to maintain the stability of the format, 1 / 2≤C<1; u n i,j ,v n i,j The velocity components of the current unit in the x and y directions, in m·s -1 ;h n i,j is the water depth of the cell, in m; g is the gravity velocity; D x and D y are the diffusion coefficients in the x and y directions, respectively, in m 2 ·s -1 ; γ is the stability coefficient of the diffusion term.
[0166] 4) Solving the land model
[0167] ① Calculation of hydrological model. First, the rainfall in the basin is calculated by inverse distance interpolation, and the infiltration is calculated according to the Green-Ampt formula. DWE is used for confluence calculation, and the flow direction of water in the unit grid is the direction with the largest water level gradient. The finite difference method is used to solve DWE:
[0168]
[0169]
[0170]
[0171] In the formula, the superscript n is the number of time steps, Q n x(i,j) ,Q n y(i,j) Land runoff in x and y directions, in m 3 ·s -1 ; S n x(i,j) ,S n y(i,j) The water level slope in the x and y directions respectively; h n i,j is the effective water depth of grid (i, j), in m; is the effective water depth of the interface on the left and right sides of the grid, in meters; Refers to the effective water depth of the upper and lower interfaces of the grid, in meters; is the water depth of the grid cell, in m, w is the side length of the rectangular grid (same value for square grid, m); n r is the Manning coefficient. The schematic diagram of the unit cell can be referred to Figure 4 .
[0172] At the end of the time step, the water depth of the grid cell is updated as follows:
[0173]
[0174] Then the water depth change within a single time step in the unit is:
[0175]
[0176] Where i is the unit grid (control volume) number; are the water depths of the unit grid (i, j) at the current moment and the next moment, in m; A i,j is the area of the unit grid; and are the inflow, outflow and source items of the grid (i, j) at the current moment, in m 3 ·s -1 .
[0177] ② Calculation of pollutant load. The process of calculating pollutant load is as follows: Figure 5 As shown. When there is no rain in a drought, pollutants accumulate on the surface; when it rains, these pollutants enriched on the surface enter the water body under the scouring of rainfall runoff, and the polluted blocks solidified on the surface become substances dissolved in the water. Based on the confluence process calculated by the hydrological model, an exponential equation is used to describe the flux of pollutants entering the water body with rainfall runoff.
[0178] ③Simulate the transport process of pollutants on land. The water quality equation is used to simulate the transport process of pollutants on land with water flow. The water quality equation is discretized using the finite volume algorithm of the Godunov center format. The discretized form of the water quality equation is as follows:
[0179]
[0180] Where i, j are the control volume numbers; m is the number of sides of the current control volume, N is the number of sides of the rectangular grid, and in the case of a uniform rectangular grid, N = 4; A i,j is the area of the (i, j) control volume. The other symbols and meanings are the same as before.
[0181] The discrete convection flux calculation method of the water quality equation can calculate the pollutant flux through the interface by judging the sign of the characteristic line based on the Riemann solution of the hyperbolic equation. The MUSCL linear reconstruction and the second-order explicit Runge-Kutta method are applied to ensure the second-order accuracy in time and space.
[0182] The MUSCL interpolation format is as follows Figure 6 As shown in . The variables are stored at the center of the grid, but not on the grid interface. However, the convective flux on the grid interface is calculated. Therefore, the MUSCL linear interpolation method is used to interpolate the variables at the center of the grid (i, j) and (i+1, j) to the interface (i+1 / 2, j). Figure 6 (c)
[0183]
[0184]
[0185] It can be seen that U i+1 / 2L ≠U i+1 / 2R
[0186] Where U represents all variables stored at the center of the grid, such as water depth, flow velocity, pollutant flux, etc.
[0187] Therefore, when solving the pollutant transport process, the MUSCL linear interpolation method is used to interpolate the variables at the center of the grid to the interface (i+1 / 2, j), and the following is obtained: n i+1 / 2-,j , z n i+1 / 2-,j , (c k ) n i+1 / 2L,j , h n i+1 / 2+,j , z n i+1 / 2+,j , (c k ) n i+1 / 2R,j and
[0188] Based on this, four characteristic lines can be calculated on the grid interface, such as Figure 7 The characteristic line is calculated as follows:
[0189]
[0190]
[0191]
[0192]
[0193] The water depth and bottom elevation in the above formula are determined by the following formula:
[0194] h i+1 / 2L,j =max(h i+1 / 2-,j +z i+1 / 2-,j -max(z i+1 / 2-,j ,z i+1 / 2+,j ),0)
[0195] h i+1 / 2R,j =max(h i+1 / 2+,j +z i+1 / 2+,j -max(z i+1 / 2-,j ,z i+1 / 2+,j ),0)
[0196] In the formula, and are the two characteristic lines on the left side of the (i+1 / 2,j) interface, and are the two characteristic lines on the right side of the (i+1 / 2,j) interface, h i+1 / 2L,j and h i+1 / 2R,j are the effective water depths on the left and right sides of the interface respectively.
[0197] The characteristic line of the (i+1 / 2,j) interface is obtained as follows:
[0198]
[0199]
[0200]
[0201] The pollutant flux through the interface (i+1 / 2,j) is calculated as
[0202]
[0203] The flux calculation in the y direction is also processed using this method.
[0204]
[0205]
[0206]
[0207]
[0208] The water depth and bottom elevation in the above formula are determined by the following formula:
[0209] h i,j+1 / 2UP =max(hi,j+1 / 2+ +z i,j+1 / 2+ -max(z i,j+1 / 2- ,z i,j+1 / 2+ ),0)
[0210] h i,j+1 / 2DN =max(h i,j+1 / 2- +z i,j+1 / 2- -max(z i,j+1 / 2- ,z i,j+1 / 2+ ),0)
[0211] In the formula, h i,j+1 / 2UP and h i,j+1 / 2DN are the effective water depths on the upper and lower sides of the (i, j+1 / 2) interface respectively; and are the two characteristic lines on the lower side of the interface (i, j+1 / 2), and are two characteristic lines on the upper side of the interface (i, j+1 / 2), such as Figure 8 shown.
[0212] Therefore, the characteristic line on the interface (i, j+1 / 2) is:
[0213]
[0214]
[0215]
[0216] Therefore, the pollutant flux through the interface (i, j+1 / 2) is calculated as
[0217]
[0218] The pollutant flux through the interface is thus obtained, and the MUSCL reconstruction method can ensure the second-order accuracy in space.
[0219] When the discrete form of the water quality equation is used to obtain the value of the variable at time (n+1), the convection term is calculated in two steps in the time direction to achieve second-order accuracy in time.
[0220]
[0221]
[0222]
[0223] Where Φ is the discrete operator of the convection term and w is the variable to be determined.
[0224] The diffusion term in the discrete form of the water quality equation is calculated as follows:
[0225]
[0226]
[0227] The water depth and flow rate used in solving the pollutant convection diffusion equation of the land model are calculated by the hydrological model in step ①. The diffusion term and the non-point source pollutant washoff calculated in step ② are added to the water quality equation as source terms, thereby obtaining the transport process of pollutants on land with water flow.
[0228] 5) Solve the water body model.
[0229] The hydrodynamic equation and water quality equation in the water body model adopt a tightly coupled solution method, that is, the hydrodynamic equation and the water quality equation are calculated simultaneously, and the variables of the hydrodynamic equation are substituted into the water quality equation for coupled solution, which can more accurately reflect the influence of hydrodynamic changes on physical processes such as water quality, which is also one of the innovative points of the present invention. The finite volume algorithm of the Godunov center format is used for spatial discretization, and the actual continuous water flow is discretized to obtain a discontinuous discontinuous distribution in the numerical calculation. The discontinuous flux is solved at each moment to obtain the input and output flux of the unit cell, and then the change of the physical quantity of the water flow (water depth, velocity, concentration) inside the unit over time is determined.
[0230] The HLLC approximate Riemann solver is used to solve the convection flux on the unit interface. First, the MUSCL linear interpolation method is used to interpolate the variables at the center points of the grid (i, j) and (i+1, j) to the interface (i+1 / 2, j), and the result is and The same method is used to process the variables in the y direction.
[0231] The vector form of the two-dimensional hydrodynamic and water quality equations coupled in the conservation mode is expressed as follows:
[0232]
[0233] Where:
[0234]
[0235]
[0236] In order to obtain a good numerical format for maintaining static equilibrium, especially in still water, a reconstruction technique was used and the calculation of fluxes at interfaces was also modified. Therefore, in the water model, the following three methods are used: the bottom slope is solved using a good equilibrium scheme (i.e., hydrostatic pressure reconstruction), the friction term is solved using a semi-implicit method, and the solution for pollutant diffusion is the same as the land model.
[0237] The purpose of the hydrostatic pressure reconstruction is to reconstruct the new variables used in the numerical fluxes to maintain the static state and the positivity of the water depth. The hydrostatic pressure reconstruction can be written as:
[0238] h i±1,j =max[0,h i,j +z i,j -max(z i,j ,z i±1,j )]
[0239] h i,j±1 =max[0,h i,j +z i,j -max(z i,j ,z i,j±1 )]
[0240] Based on the hydrostatic reconstruction technique, the flux calculation is corrected. In this case, the flux gradient in the numerical scheme must be balanced with the approximation of the source term. To analyze this, only the bottom slope in the source term is considered, because the friction term does not affect the hydrostatic solution. Therefore, the bottom slope and the corrected flux are calculated first. The coupled form of the two-dimensional hydrodynamic and water quality equations can be rewritten as follows:
[0241]
[0242] In the formula,
[0243] Discretize the linear equations and obtain the calculation formula of the homogeneous solution with second-order time and space accuracy:
[0244]
[0245] in:
[0246]
[0247]
[0248]
[0249]
[0250]
[0251]
[0252] In the formula, h i+1 / 2+,j ,h i+1 / 2-,j are the water depths on the left and right sides of the interface (i+1 / 2,j); h i+1 / 2L,j ,h i+1 / R,j are the effective water depths on the left and right sides of the (i+1 / 2,j) interface, h i,j+1 / 2Rand h i,j+1 / 2L are the effective water depths on the upper and lower sides of the (i, j+1 / 2) interface, respectively, and the calculation method is the same as before; S is the correction term, the purpose of which is to ensure hydrostatic balance; Sc is the center value, the purpose of which is to ensure the second-order accuracy of hydrostatic balance.
[0253] Due to the particularity of shallow water flow, the four variables of F and G are calculated using different methods. 2 +gh 2 / 2,huv,huc) the first two variables (hu,hu 2 +gh 2 / 2) and G(hv,huv,hv 2 +gh 2 / 2,hvc)'s first and third variables (hv, hv 2 +gh 2 / 2), which can be calculated using the HLLC approximate Riemann solver.
[0254] The flux in the x-direction is calculated as follows:
[0255] The velocity and water depth at time n calculated using the two-dimensional hydrodynamic model give four characteristic lines on the left and right sides of the (i+1 / 2,j) interface: and like Figure 7 As shown. The flux through the interface (i+1 / 2,j) is obtained as follows:
[0256]
[0257] Similarly, the flux in the y direction is calculated as follows:
[0258] The velocity and water depth at time n calculated using the two-dimensional hydrodynamic model are used to calculate four characteristic lines on the upper and lower sides of the (i, j+1 / 2) interface: and like Figure 8 As shown. The flux through the interface (i+1 / 2,j) is obtained as follows:
[0259]
[0260] The third and fourth variables of F (huv, huc k ) and the second and fourth variables of G (huv, hvc k ) can be directly determined by the first variable (hu, hv) of F and G, and the calculation method is as follows:
[0261]
[0262]
[0263]
[0264]
[0265] When calculating the value of the variable at time (n+1), the convection term is calculated in two steps in the time direction to achieve second-order accuracy in time.
[0266] Solve for the friction and diffusion terms:
[0267] A semi-implicit method is used to solve the friction source term, which effectively ensures the stability of the algorithm.
[0268]
[0269] The treatment of diffusion terms in the water quality equation is the same as that in the land model. Unlike the water quality model on land, the water quality equation in the water body does not calculate the non-point source pollution flushing source term, but only considers the degradation of pollutants caused by the self-purification capacity of the water body.
[0270] 6) Update the calculation range of the next moment's flooded area and non-flooded area according to the obtained water depth h, that is, when h is greater than the preset water depth threshold, it is regarded as a flooded area and calculated using the water body model; when h is less than the threshold, it is regarded as a non-flooded area and calculated using the land model. Repeat steps 2) to 5) until the calculation is completed.
[0271] like Figures 9 to 14 The performance of the model was verified by simulating the movement of pollutants on a two-dimensional slope and verifying the coupling effect between the slope hydrological area and the flooded area with greater water depth. The computational domain is 500 meters long and 400 meters wide, with a slope of S in the x direction. x =0.02+0.0000149x, the slope in the y direction is S x =0.02+0.0000116y. Roughness Where n rx =0.1-0.0000168x,n ry =0.1-0.0000168y. The rainfall in this example changes linearly with time. At t=0s, the rainfall intensity is 0mm / s. At t=6000s, the rainfall intensity reaches a maximum of 28.8mm / h. At t=12000s, the rainfall intensity drops linearly to 0mm / s. The total simulation time is 14400s. The calculation area is discretized using a uniform rectangular grid, such as Fig.10 As shown in the figure. The initial conditions are that the water depth and flow velocity in the entire area are 0, and the boundary conditions are that the two edges of x=0 and y=0 use solid wall boundaries, and the other two edges are free outflow boundaries. The initial pollution condition is set in a circular area in the upper right corner of the area, with the center at (180, 230) and the radius at 5m, as shown in the figure. Fig.11 As shown. The mass of the pollutant is 10g / m 2 , given a diffusion coefficient of 0.5 m 2 / s.
[0272] First, the flow rate at the outlet is compared with the results of other models to verify the rationality of the model. The results are compared with the numerical model developed by Jaber et al. and the two-dimensional hydrodynamic model developed by IFMS developed by China Institute of Water Resources and Hydropower Research. Fig.12 The results obtained by this model are close to those of the IFMS of the Institute of Water Resources and Hydropower Research and the two-dimensional hydrodynamic model of Jaber et al., indicating that the coupling model of this invention can better deal with the problems of rainfall runoff and the planar evolution of water flow in the terrain.
[0273] Fig.13 The pollutant concentration distribution diagram at two moments of the watershed water environment model of the present invention for the simulation calculation of pollution generation and sink for the example of variable slope and variable roughness. The numerical simulation shows that the pollutants on the slope move downstream along the slope with the scouring of rainfall runoff. The diffusion phenomenon of pollutants can also be found from the figure. The concentration of the center position of the pollutant fluid group decreases with the transfer process to the downstream of the basin. The direction of water flow movement upstream of the ramp points to the downstream, but the lateral diffusion process of pollutants to the ramp can be clearly seen. The diffusion process increases with the increase of the turbulence intensity of the water flow, making the molecular movement of pollutants more intense, so in Fig.13 In (b), it can be seen that the concentration of pollutants is more widely distributed.
[0274] In order to better illustrate the conservation properties of the watershed water environment model of the present invention, the mass of pollutants remaining in the watershed and the mass of pollutants discharged from the watershed at different times in the example are statistically analyzed. Fig.14 As shown in the figure, the numerical calculation of the present invention has good conservation. As the rainfall intensity increases, the pollutants flow out of the basin mainly between 6000s and 8000s. Before 6000s, the rainfall intensity has not reached the peak, and it takes time for the pollutant fluid group to move to the outlet position. After 8000s, there are fewer pollutants left in the basin and the concentration of pollutants is low.
[0275] It can be seen from the above specific implementation methods that the basin water environment simulation method and device provided by the present invention overcomes the shortcomings of using a single land model or water model and a unidirectional coupling of a land model and a water model to predict basin water environment pollution by directly and dynamically bidirectionally coupling a land model with a water model. By judging the state of the grid according to a preset water depth threshold, the area with a deeper water depth is regarded as a flooded area, and the convection diffusion process of pollutants occurs. The water body model is used to calculate the water depth and flow rate of the flooded area, and the concentration of pollutants is solved. The area with a shallower water depth is regarded as a non-flooded area, and the transfer process of pollutants occurs. The land model is used to simulate the generation and transfer process of pollutants on land. The dynamic bidirectional coupling of the two models established does not specify the boundary position in the calculation. The location where pollutants and water flow enter the flooded area is dynamically changing. The flood propagation can also flood the original confluence point. The boundary shift forms a new coupling boundary, and the alternating conversion of the two areas is realized, and the simulation is closer to the real situation. Compared with the external single coupling method, which requires the calculation of the land model and the water model in sequence, the two models are calculated simultaneously in the bidirectional coupling method adopted by the present invention. Pollutants can flow from the non-flooded area into the flooded area, and pollutants in rivers and lakes can also invade the land area of water collection, thereby realizing the mutual influence of pollutants on the shore and in the water body, which is more in line with physical reality and more scientific. The coupled moving boundary is used as the boundary of the control volume unit, and different calculation models (land model and water body model) are used on both sides of the moving boundary. Through the finite volume format of water volume, momentum, and pollutant mass conservation, it not only ensures the conservation of various physical quantities, but also can automatically realize the land pollutants entering the water body through the coupled moving boundary, realizing the driving of the water model by the land model, instead of setting boundary conditions artificially and loading land pollutants as boundary conditions into the water model.
[0276] As described above, the method and device for simulating the water environment of a river basin proposed by the present invention are described by way of example with reference to the accompanying drawings. However, it should be understood by those skilled in the art that various improvements can be made to the method and device for simulating the water environment of a river basin proposed by the present invention without departing from the content of the present invention. Therefore, the protection scope of the present invention should be determined by the content of the attached claims.
Claims
1. A method for simulating a watershed water environment, characterized in that: The steps include: S1. According to a preset water depth threshold, the watershed to be predicted is divided into grids, and the grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the grids with water depths greater than the preset water depth threshold are regarded as flooded grids; S2. Connecting a land model and a water body model by using the boundary between the non-flooded area formed by the non-flooded grid and the flooded area formed by the flooded grid as a coupled dynamic boundary; wherein the land model includes: a hydrological model for calculating the convergence velocity and water depth of the non-flooded grid, a non-point source pollution load model for calculating the accumulation amount and flushing amount of pollutants in the non-flooded grid according to the convergence velocity and water depth of the non-flooded grid, and a pollutant transport model for obtaining non-point source pollutant load distribution data by taking the accumulation amount and flushing amount of pollutants in the non-flooded grid as source items; The water body model includes: a hydrodynamic model for obtaining the flow velocity and water depth in the submerged grid by solving the two-dimensional shallow water equation in the local area using the confluence velocity and water depth of the non-submerged grid as the driving conditions for the water flow movement of the water body, and a water quality model for solving the convection diffusion process of pollutants in the water body based on the flow velocity and water depth in the submerged grid using the concentration distribution data of non-point source pollutants in the non-submerged grid as the driving conditions for the convection diffusion process of pollutants in the water body; S3. For the non-flooded grid, the land model is used to simulate the generation of pollutants on land and the transport process with runoff, and the pollutant load distribution data of the non-flooded area is calculated; for the flooded grid, the pollutant load distribution data of the non-flooded area is used as a calculation condition, and the water body model is used to simulate the convection diffusion process of pollutants in the water body, and the concentration distribution data of pollutants in the flooded area is calculated; wherein, in the process of using the water body model to simulate the convection diffusion process of pollutants in the water body for the flooded grid, and using the pollutant load distribution data of the non-flooded area as a calculation condition to calculate the concentration distribution data of pollutants in the flooded area, The hydrodynamic model and the water quality model are coupled and solved by a tightly coupled solution method, and the variable at the center point of the submerged grid is interpolated to the interface of the submerged grid by a MUSCL linear interpolation method; The Godunov format finite volume method is used for spatial discretization, the HLLC approximate Riemann solver is used to solve the convective flux on the submerged grid interface, the second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, and the water depth and flow velocity of the grid interface of the submerged area are solved; Based on the conservation of mass and momentum of the coupled moving boundary interface, the pollutant load distribution of the non-flooded area is used as the calculation condition of the water body model, and the water depth and flow velocity of the flooded grid are substituted into the water quality equation as known quantities, the Godunov format is used for spatial discretization, the HLLC approximate Riemann solution is used to solve the flux of pollutants passing through the flooded grid interface, the second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, so as to obtain the concentration distribution data of the pollutants in the flooded area; S4. Calculate the water depth data of the submerged grid and the non-submerged grid based on the time step of model calculation, and update the coupled moving boundary according to the water depth data, and return to step S2.
2. The basin water environment simulation method according to claim 1, characterized in that: According to the preset water depth threshold, the watershed to be predicted is divided into grids, and the grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the grids with water depths greater than the preset water depth threshold are regarded as flooded grids, including: Dividing the watershed to be predicted into uniform rectangular grids; Defining the boundaries of the rectangular grids and the relationship between adjacent rectangular grids, and numbering the rectangular grids based on the definitions; The variables are stored in the center position of the numbered rectangular grids, initial conditions are set for each numbered rectangular grid, and the numbered rectangular grids with water depths lower than the preset water depth threshold are regarded as non-flooded grids, and the numbered rectangular grids with water depths greater than the preset water depth threshold are regarded as flooded grids.
3. The basin water environment simulation method according to claim 1, characterized in that: In the process of using the land model to simulate the generation of pollutants on land and the transport process of pollutants with runoff for the non-flooded grid and calculating the pollutant load distribution data of the non-flooded area, In the hydrological model: The rainfall of the non-flooded grid watershed is calculated by the inverse distance interpolation method, and the infiltration of the watershed is calculated by the Green-Hampton formula; Calculating the net rainfall of the non-flooded grid based on the rainfall and infiltration of the watershed; The diffusion wave equation is used to calculate the convergence velocity and water depth of the non-flooded grid; In the non-point source pollution load model, based on the convergence velocity and water depth calculated by the hydrological model, an exponential equation is used to calculate the amount of pollutants entering the water body with rainfall runoff; In the pollutant transport model: The water quality equation is used to simulate the transport process of pollutants on land with water flow, and the water quality equation is discretized using a finite volume algorithm of the Godunov center format to obtain the pollutant load distribution data of the non-flooded area.
4. The basin water environment simulation method according to claim 1, characterized in that: The method for determining the time step of the model calculation includes: The time step of the model calculation is determined by using an explicit grid discretization method; wherein the time step constraint is: in Where C is a constant used to ensure the stability of the format, 1 / 2≤C<1; u n i,j ,v n i,j are the velocity components of the current unit in the x and y directions (m·s -1 );h n i,j is the water depth of the unit grid (m); g is the gravity velocity; D x and D y are the diffusion coefficients in the x and y directions (m 2 ·s -1 ), γ is the stability coefficient of the diffusion term.
5. The basin water environment simulation method according to claim 1, characterized in that: Calculating water depth data of the submerged grid and the non-submerged grid based on a time step of model calculation, and updating the coupled moving boundary according to the water depth data, including: calculating characteristic waves at discontinuities between the submerged grid and the non-submerged grid, and determining the direction of mass and momentum transfer between the submerged area and the non-submerged area according to the direction of the characteristic waves; updating the water depth and flow velocity of the non-flooded area and the flooded area according to the mass and momentum transfer direction between the flooded area and the non-flooded area; According to the water depths of the non-flooded area and the flooded area, taking the grids whose water depths are lower than the preset water depth threshold as updated non-flooded grids, and taking the grids whose water depths are greater than the preset water depth threshold as updated flooded grids; The coupled moving boundary connecting the land model and the water body model is updated according to the boundary between the new non-flooded area formed by the updated non-flooded grid and the new flooded area formed by the updated flooded grid.
6. The method for simulating water environment in a watershed according to claim 5, characterized in that: The calculating of the characteristic waves at the discontinuities between the submerged grid and the non-submerged grid, and determining the mass and momentum transfer direction between the submerged area and the non-submerged area according to the direction of the characteristic waves comprises: When the directions of the characteristic waves at the coupled moving boundary all point to the submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the non-submerged grid, and it is determined that the non-submerged area transfers mass to the submerged area but does not transfer momentum; When the characteristic wave at the coupled moving boundary is distributed in both the submerged grid and the non-submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the submerged grid and the water depth and flow velocity of the non-submerged grid, and it is determined that the non-submerged area transfers mass to the submerged area and the submerged area transfers momentum to the non-submerged area; When the directions of the characteristic waves at the coupled moving boundary all point to the non-submerged grid, the flux at the discontinuity between the submerged grid and the non-submerged grid is calculated according to the water depth and flow velocity of the submerged grid, and it is determined whether the submerged area transfers mass and momentum to the non-submerged area.
7. A watershed water environment simulation device, characterized in that: include: A grid division module is used to divide the watershed to be predicted into grids according to a preset water depth threshold, and to divide the grids whose water depth is lower than the preset water depth threshold into non-flooded grids, and the grids whose water depth is greater than the preset water depth threshold into flooded grids; A model building module, used to build a land model and a water body model with the boundary between the non-flooded area formed by the non-flooded grid and the flooded area formed by the flooded grid as a coupled dynamic boundary; wherein the land model includes: a hydrological model for calculating the convergence velocity and water depth of the non-flooded grid, a non-point source pollution load model for calculating the accumulation and flushing amount of pollutants in the non-flooded grid according to the convergence velocity and water depth of the non-flooded grid, and a pollutant transport model for obtaining non-point source pollutant load distribution data by taking the accumulation and flushing amount of pollutants in the non-flooded grid as source items; The water body model includes: a hydrodynamic model for obtaining the flow velocity and water depth in the submerged grid by solving the two-dimensional shallow water equation in the local area using the confluence velocity and water depth of the non-submerged grid as the driving conditions for the water flow movement of the water body, and a water quality model for solving the convection diffusion process of pollutants in the water body based on the flow velocity and water depth in the submerged grid using the concentration distribution data of non-point source pollutants in the non-submerged grid as the driving conditions for the convection diffusion process of pollutants in the water body; The calculation module is used to use the land model to simulate the generation of pollutants on land and the transport process with runoff for the non-flooded grid, and calculate the pollutant load distribution data of the non-flooded area; for the flooded grid, the pollutant load distribution data of the non-flooded area is used as a calculation condition, and the water body model is used to simulate the convection diffusion process of pollutants in the water body, and calculate the concentration distribution data of pollutants in the flooded area; wherein, in the process of using the water body model to simulate the convection diffusion process of pollutants in the water body for the flooded grid, and using the pollutant load distribution data of the non-flooded area as a calculation condition to calculate the concentration distribution data of pollutants in the flooded area, The hydrodynamic model and the water quality model are coupled and solved by a tightly coupled solution method, and the variable at the center point of the submerged grid is interpolated to the interface of the submerged grid by a MUSCL linear interpolation method; The Godunov format finite volume method is used for spatial discretization, the HLLC approximate Riemann solver is used to solve the convective flux on the submerged grid interface, the second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, and the water depth and flow velocity of the grid interface of the submerged area are solved; Based on the conservation of mass and momentum of the coupled moving boundary interface, the pollutant load distribution of the non-flooded area is used as the calculation condition of the water body model, and the water depth and flow velocity of the flooded grid are substituted into the water quality equation as known quantities, the Godunov format is used for spatial discretization, the HLLC approximate Riemann solution is used to solve the flux of pollutants passing through the flooded grid interface, the second-order Runge-Kutta method is used to ensure the second-order accuracy of the time integration, and the MUSCL format is used to ensure the second-order accuracy in space, so as to obtain the concentration distribution data of the pollutants in the flooded area; The grid updating module is used to calculate the water depth data of the submerged grid and the non-submerged grid based on the time step of the model calculation, and update the coupled moving boundary according to the water depth data.
