CN102722909B - Assembly line topology network dynamic simulation method based on adaptive-dimensional DEM (dynamic effect model) - Google Patents

Abstract

The present invention proposes a dynamic simulation method of pipeline topology network based on self-adaptive scale DEM, that is to make full use of the increasing high-precision DEM data, extract terrain and hydrological features, and assign scale attributes to the features, thereby constructing a TIN based on river network constraints The Adaptive Scale DEM Database. On this basis, the water flow direction of each triangle in the TIN is calculated, and the water flow line from any position to the outlet of the watershed is constructed, thereby simplifying the three-dimensional terrain expression into a one-dimensional topological pipeline structure, and realizing nested multi-scale and multi-level Topological structure, using classical hydrological formulas to calculate relevant hydrological parameters, and realize multi-scale applications of dynamic simulation of surface water flow from wide area to local area.

Description

A Dynamic Simulation Method of Pipeline Topological Network Based on Adaptive Scale DEM

technical field

The invention belongs to the field of digital terrain analysis, in particular to a dynamic simulation method of pipeline topology network based on self-adaptive scale DEM.

Background technique

The dynamic simulation and prediction of geoscience processes is a research hotspot in the fields of geography, environmental science, information science and so on. As a typical application of dynamic simulation of geoscience process, hydrological model has attracted the attention of many experts and scholars for a long time. For example, through the dynamic simulation of the surface water process, the flood disaster situation can be predicted on the basis of the rainfall runoff formation principle and the flood wave movement law, so as to assist in flood control decision support; the pollution caused by sudden water pollution accidents can be simulated , so that the affected scope and objects can be determined in time, and appropriate measures can be taken to control its adverse effects.

The scale problem in spatiotemporal dynamic simulation involves two aspects: spatial scale and time scale. Many terrain parameters related to hydrology are continuously changing, and the terrain must be simplified during the digitization process. The higher the spatial scale (spatial resolution), the better the digital terrain can express the real hydrological parameters, but the spatial resolution increases, and the data storage And the larger the number of handles. Similarly, in the time scale, the shorter the time interval between the recording and calculation of hydrological parameters, the more realistic the description of the hydrological process, but it will greatly increase the amount of storage and calculation. On the other hand, different hydrological processes may act on different spatial and temporal scales, so accurately describing these hydrological processes requires hydrological models to be compatible with multiple time scales at the same time. How to adaptively select the required scale according to the hydrological process is a key problem to be solved urgently.

During decades of research, the hydrological model has experienced the development process from the traditional lumped model, to the semi-distributed model, and then to the spatially distributed model (Saint-Venant, 1871; Abbott et al., 1986; Turcotte et al., 2001). The current hydrological model needs to fully consider the environmental factors and the spatial and temporal distribution of hydrological processes, while the digital terrain modeling method effectively provides management and analysis methods for geospatial information, making it possible to predict the temporal and spatial patterns of runoff based on physical principles under certain conditions become possible.

The terrain surface description method usually used in hydrological models is regular grid DEM (Li Zhilin et al., 2000; Zhou Qiming et al., 2006). DEM stands for Digital Elevation Model. With the emergence of early GIS (Geographic Information System), hydrological models based on regular grid DEM have been developed and widely used in distributed hydrological models. Beven et al. (1979) proposed a terrain-based hydrological model (TOPMODEL). In order to calculate the production and confluence, the model calculated topographic parameters based on DEM to reflect the hydrological characteristics of the surface and shallow underground layers. However, TOPMODEL does not consider hydrological physical processes such as rainfall and evaporation, so it only partially simulates hydrological processes. The SHE model (System Hydrologic European) jointly developed and improved by Beven et al. (1980) is a typical distributed hydrological model, and its watershed is divided into three-dimensional (vertical multi-layer) regular grids to combine model parameters and rainfall input to simulate hydrological processes. Arnold (1994) developed the SWAT model (Soil and Water Assessment Tool) for USDA. The SWAT model belongs to the semi-distributed hydrological model, which uses grid DEM to integrate the spatial information provided by GIS and RS (remote sensing) to divide the hydrological feature area to reduce the amount of calculation, thereby simulating the hydrophysical process in complex large watersheds.

Since the regular grid DEM itself is an approximate description of the continuous surface in the real world using regular space sampling, its accuracy is restricted by factors such as measurement error, sampling error, and grid resolution, and it is difficult to fully and accurately express various terrain features. Although the finer the scale, the closer the described spatiotemporal process is to real data, but it also greatly increases the amount of storage and calculation. In order to make the selected DEM data scale meet the scale required for hydrological simulation, the DEM data synthesis method is usually adopted (Fei Lifan et al., 2006, Hu Peng et al., 2009). Zhang and Montgomery (1994) used different grid resolutions (2-90m) to test the influence of grid size on terrain representation and hydrological simulation. The results showed that DEM grid resolution has a great influence on terrain parameters and water level calculation. Zhou and Liu (2004) studied the influence of DEM raster resolution on various slope and aspect algorithms, and concluded that the uncertainty of DEM terrain parameters is closely related to the raster data structure.

Another approach is to use triangulated irregular network (TIN) models in hydrological applications. TIN has certain vector data characteristics, can effectively express points, lines and surfaces of any size, shape and angle, and can easily embed feature points ("turning points", etc.) and feature lines (pipeline, ridge line, etc.), so that Effectively express terrain changes. Vivoni, et al. (2005) embedded hydrological characteristics such as river lines, river boundaries, and floodplains in the TIN model to construct a digital surface model suitable for hydrological applications. Zhou and Chen (2011) proposed a terrain compound point extraction algorithm (Compound Point Extraction, CPE) to construct the TIN structure subject to pipeline constraints. The direction of maximum slope delineates the flow path.

Compared with the hydrological model based on regular grid DEM, the hydrological model based on TIN terrain representation has some natural advantages in some aspects. However, due to the irregularity of the TIN data structure, it poses great challenges to the simulation of spatial hydrological processes and related algorithms. The comprehensive method based on TIN mainly uses characteristic points (such as inflection points) and characteristic lines (such as watershed lines) to construct TINs to express topographic surfaces at different scales in hydrological applications (Heller, 1990). Kidner et al. (2000) proposed a topology-free TIN structure for multi-scale data models, which only store feature points and lines (such as watershed lines and ridge lines), and constructed a hierarchical TIN model that facilitates multi-scale query. Danovaro et al. (2006) proposed a multi-resolution surface network (MSN), using feature points (minimum point, maximum point and saddle point) and feature lines to construct TIN to describe the surface of various resolutions.

Whether it is based on grid DEM or TIN, the expression of surface model only describes the terrain surface, and cannot show the dynamic characteristics of surface water flow. Since surface water is the most active factor in the natural geographical environment, the dynamic simulation of surface water process is also the most important part of geoscience process simulation. The key issue in the simulation and prediction of surface water dynamics is how to determine the relationship between flow and velocity (Djokic and Maidment, 1993).

Early simulations of surface water dynamics used empirical formulas to calculate (Dietrich et al, 1993). With the development of digital terrain model and digital terrain analysis research, it is possible to use physical models to predict runoff and soil erosion under certain environmental conditions (Beven and Moore, 1994). Bates and Roo (2000) proposed a one-dimensional kinematic wave approximation for river runoff and a two-dimensional diffuse wave representation for floodplain runoff. Tucker et al. (2001) proposed the CHILD model (Channel-Hillslope Integrated Landscape Development), which simulates terrain changes caused by erosion and deposition based on the TIN model. Vivoni et al. (2005) developed the tRIBS model (TIN-based Real-time Integrated Basin Simulator) to predict the hydrological response of the surface and subsurface after rainfall.

Despite the efforts of scientists for many years, most of the currently applied hydrological models are non-distributed or semi-distributed hydrological models, and GIS is only used to calculate the hydrological parameters of the watershed. The hydrological models actually used in engineering are mainly non-distributed hydrological models bound to geographic information systems using traditional hydrological models, such as TR20, HEC-1, SWAT, etc. Distributed hydrological models are basically only used for small-scale laboratory research Hydrological dynamics. This is because the distributed hydrological model has not been able to properly resolve the contradiction between the resolution of time and space scales and the huge amount of calculation.

Most of the existing inventions use grid-based distributed hydrological models to perform single-scale hydrological simulations and flood forecasts, and are related to:

Zhang Wanchang et al. (2011, see background document 23) of Nanjing University proposed a distributed hydrological model design method using grid as the simulation unit. It converts the distributed parameter vector data into raster data, establishes the common runoff type of the grid, and performs the calculation of the runoff and confluence process of the model under different conditions through the design of the runoff and confluence process. Finally, the hydrological process simulation of the watershed in the arid and humid areas, as well as the short-term flood forecast and the long-term rainfall-runoff process simulation of the watershed will be realized.

Ran Qihua et al. (2011, see Background Document 24) from Zhejiang University proposed a flood forecasting method based on rainfall-runoff-flood evolution calculations. It integrates hydrological data according to the specifications and requirements of the distributed hydrological model and hydrodynamic model, and uses the model to carry out the flood evolution process; compares and calculates the water level forecast data and warning water level data of key nodes in the river basin, and releases the results. In the end, the flood forecast of each point can be conveniently carried out according to the measured rain and water regime.

Li Chunhong et al. (2010, see Background Document 25) from the State Grid Electric Power Research Institute proposed a hydrological forecasting method combining hydrological models with different mechanisms. According to the characteristics of the watershed, configure three or more medium-term hydrological forecast models that conform to the characteristics of the watershed, and use historical hydrological data to determine the model parameters with the highest comprehensive accuracy for each hydrological forecast model rate. At each forecast moment, configure different forecast combination schemes, and perform pre-test forecast calculations based on optimal parameters; automatically evaluate the test forecast calculation results of each combination scheme, obtain the current optimal combination method, and finally apply it to the current hydrological forecast.

The above inventions do not take into account the problem of time and space scales, and it is difficult to support complex hydrological models of watersheds. In order to meet the multi-scale or cross-scale application requirements from wide area to local scale, new theories and methods are needed to improve the accuracy, scope and efficiency of dynamic simulation of surface water flow.

Background literature:

1. Li Zhilin, Zhu Qing, 2000, Digital Elevation Model [M], Wuhan: Wuhan University of Surveying and Mapping Press.

2. Zhou Qiming, Liu Xuejun, 2006, Digital Terrain Analysis, Beijing: Science Press.

3. Hu Peng, Gao Jun, 2009, Research on Digital Synthesis Principle of Digital Elevation Model, Journal of Wuhan University Information Science Edition, 34(8): 940-942.

4. Fei Lifan, He Jin, Ma Chenyan, Yan Huiwu, 2006, 3D Douglas-Peucker algorithm and its application in DEM automatic synthesis, Journal of Surveying and Mapping, 35 (8): 278-284.

5.Abbott,M.B.,Bathurst,J.A.,Cunge,P.E.,1986.An introduction to the European hydrologicalsystem-systeme hydrologique Europeen "SHE":part 1.History and philosophy of a physically based distributed modeling system.Journal 4y of 8 Hydrology,7 59.

6. Arnold, J.G., Williams, J.R., Srinivasan, R., King, K.W., Griggs, R.H., 1994. SWAT-Soil and Water Assessment Tool-User Manual. Agricultural Research Service, Grassland, Soil and Water Research Laboratory, US Department of Agriculture .

7. Bates, P.D., De Roo, A.P.J., 2000. A simple raster-based model for flood inundation simulation. Journal of Hydrology 236, 54-77.

8. Beven, K.J., Kirkby, M.J., 1979. A physically based, variable contributing area model of basinhydrology. Hydrological Sciences Bulletin 24, 43-69.

9. Beven, K.J., Moore, I.D., 1993. Terrain Analysis and Distributed Modeling in Hydrology. Chichester, UK: John Wiley&Sons, 249pp.

10. Beven, K.J., Warren, R., Zaoui, J., 1980. SHE: Towards a Methodology for Physically-Based Distributed Forecasting in Hydrology. In Hydrological Forecasting, IAHS Publication 129, 133-137.

11. Danovaro, E., Floriani, L.D., Papaleo, L, 2006. A multi-resolution representation for terrain morphology. Lecture Notes in Computer Science 4197:33-46.

12. Dietrich, W.E, Wilson, C.J., Montgomery, D.R. and McKean, J., 1993. Analysis of erosionthresholds, channel networks, and landscape morphology using a digital terrain model. Journal of Geology 101, 259-278.

13. Djokic, D., Maidment, D.R., 1993. Application of GIS network routines for water flow and transport. Journal of Water Resources Planning and Management 119(2), 229-245.

14. Heller, M., 1990. Triangulation algorithms for adaptive terrain modeling. Proc. 4th International Symposium on Spatial Data Handling, Zürich, 1:163-174.

15.Kidner,D.B.,Ware,J.M.,Sparkes,A.J.,Jones,C.B.,2000.Multiscale terrain and topographic modeling with the implicit TIN.Transactions in GIS 4(4):379-408.

16.Saint-Venant, Barre de, 1871. Theory of unsteady water flow, with application to river floods and to propagation of tides in river channels. French Academy of Science 73, 148-154, 237-240.

17. Tucker, G.E., Lancaster, S.T., Gasparini, N.M., Bras, R.L., 2001. The Channel-Hillslope Integrated Landscape Development (CHILD) model. In Harmon, R.S., Doe, W. (Ed.), Landscape Erosion and Evolution Modeling , Dordrecht, Kluwer Academic/Plenum Publishers, pp.349–388.

18. Turcotte, R., Fortin, J.P., Rousseau, A.N., Massicotte, S., Villeneuve, J.P., 2001. Determination of the drainage structure of a watershed using a digital elevation model and a digital river and lake network. Journal of 2 5–40 Hydrology, 2 242.

19.Vivoni, E.R., Teles, V., Ivanov, V.Y., Bras, R., Dara, E., 2005. Embedding landscape processes into triangulated terrain models. International Journal of Geographical Information Science 19(4), 429-457.

20. Zhang, W., Montgomery, D.R., 1994. Digital elevation model grid size, landscape representation, and hydrologic simulations. Water Resources Research 30(4): 1019-1028.

21.Zhou,Q.,Chen,Y.,2011.Generalization of DEM for terrain analysis using a compound method.ISPRS Journal of Photogrammetry and Remote Sensing 66(1),38-45.

22. Zhou, Q., Liu, X., 2004. Error analysis on grid-based slope and aspect algorithms, Photogrammetric Engineering and Remote Sensing 70(8), 957-962.

23. Nanjing University, Zhang Wanchang, Zhang Dong, A Distributed Hydrological Model Design Method Using Grid as Simulation Unit, Chinese Patent, 201010590169, 2011-04-27.

24. Zhejiang University, Ran Qihua, Wang Zhenyu, He Zhiguo, Flood Forecasting Method Based on Rainfall-Runoff-Flood Evolution Calculation, Chinese Patent, 201110207840, 2011-12-21.

25. State Grid Electric Power Research Institute, Li Chunhong, Wang Feng, Zhang Jun, Lv Zhongcheng, A Hydrological Forecasting Method Combining Hydrological Models with Different Mechanisms, 200910234628, 2010-11-03.

Contents of the invention

What the present invention aims to solve is the multi-scale application requirements of hydrological simulation under massive data, and provides a brand-new pipeline topology network model (Topological Flow-path Network model) based on adaptive scale DEM (Scale-adaptive DEM, referred to as: S-DEM) , referred to as: TFN) to realize the dynamic simulation of surface water flow.

The dynamic simulation method of the pipeline topology network based on the self-adaptive scale DEM provided by the technical solution of the present invention comprises the following steps:

Step 1, based on the self-adaptive scale DEM data structure, generate the corresponding irregular triangular network according to the specified scale information;

Step 2, calculating the water flow direction line of each triangular surface in the irregular triangular network obtained in step 1, constructing a water flow path from any position to the outlet of the basin, and establishing a pipeline topology network according to the obtained water flow path;

Step 3, based on the pipeline topology network obtained in step 2, perform dynamic simulation of surface water flow.

Moreover, step 2 includes the following sub-steps,

Step 2.1, divide the irregular triangular network into regular grids, and randomly select a sampling point in each grid as the rainfall source point;

Step 2.2, by calculating the slope and aspect of each triangular surface on the irregular triangular network, the water flow direction line of each triangular surface is obtained, starting from any rainfall source point, tracing according to the water flow direction line, and constructing from any position to the water flow path at the outlet of the watershed;

In step 2.3, a pipeline topology network is established based on the water flow path obtained in step 2.2.

Moreover, in step 3, the dynamic simulation of surface water flow is carried out under the condition that only the gravity factor is considered, and the realization method is as follows, to simulate the rainfall process of uniform and non-uniform distribution in time and space over a period of time, according to the topological network of the pipeline obtained in step 2, using the hydrological formula Calculate the flow velocity of each streamline segment in the pipeline topology network, and obtain the rainfall-runoff simulation curve at any point in the rainfall area at any time period.

Moreover, in step 3, the dynamic simulation of surface water flow is carried out in consideration of gravity factors and other environmental variables, and the implementation method is as follows,

Using the real rainfall data, the hydrological model is used to calculate the runoff data, and according to the pipeline topology network obtained in step 2, the confluence results at different times are dynamically simulated.

The present invention proposes a dynamic simulation method of pipeline topology network based on self-adaptive scale DEM, that is, making full use of the increasing high-precision DEM data, extracting terrain and hydrological features, and assigning scale attributes to the features, thereby constructing a TIN based on river network constraints The Adaptive Scale DEM Database. On this basis, the water flow direction of each triangle in the TIN is calculated, and the water flow line from any position to the outlet of the basin is constructed, thereby simplifying the three-dimensional terrain expression into a one-dimensional topological pipeline structure, and realizing nested multi-scale and multi-level Topological structure, using classical hydrological formulas to calculate relevant hydrological parameters (such as velocity, flow), and realize multi-scale applications of dynamic simulation of surface water flow from wide area to local area.

Description of drawings

Fig. 1 is a flowchart of an embodiment of the present invention.

Fig. 2 is a flow chart of adaptive scale DEM modeling according to an embodiment of the present invention.

FIG. 3 is a flow chart of establishing a pipeline topology network according to an embodiment of the present invention.

Fig. 4 is a schematic diagram of a dynamic simulation of surface water flow according to an embodiment of the present invention.

Fig. 5 is a flow chart of the accuracy evaluation of the embodiment of the present invention.

Specific implementation method

The core problem to be solved by the present invention is to construct a pipeline topology network model based on self-adaptive scale DEM, carry out dynamic simulation of water flow at multiple scales, and maintain the accuracy and consistency of the results at different scales, so as to meet the needs of wide-area Multi-scale or cross-scale application requirements to a local scale, while improving the accuracy, scope and efficiency of surface water flow dynamic simulation.

Referring to Fig. 1 for the flow of the embodiment, a pipeline topology network (TFN) model based on adaptive scale DEM (S-DEM) is adopted to realize dynamic simulation of surface water flow. Feature points and lines are extracted from high-precision DEM data, and an adaptive scale DEM database (S-DEM) based on river network constraints TIN is constructed. On this basis, calculate the water flow direction of each triangle in the TIN, and construct the water flow line from any position to the outlet of the watershed, thereby simplifying the three-dimensional terrain expression into a one-dimensional pipeline topology (TFN), and establishing the topology between water flow lines relationship, realize nested multi-scale and multi-level topology structure, and use classical hydrological formulas to calculate relevant hydrological parameters (such as velocity, flow rate), and perform dynamic simulation of surface water flow based on pipeline topology structure, and finally evaluate pipeline topology at different scales The simulation accuracy of the network is used to verify the accuracy of the simulation results. The following sub-steps describe the specific implementation process of the embodiment in detail:

Step 1: Based on the self-adaptive scale DEM data structure, the corresponding irregular triangulation network is generated according to the specified scale information.

Topographic feature points and lines can be extracted from a fine-scale DEM database in advance, the mapping relationship between topographic feature points, lines and scales can be constructed, and scale information attributes can be assigned to topographic features, thereby constructing an S-DEM topographic feature database. When implementing the technical solution of the present invention, according to the application requirement scale, feature points and lines satisfying the scale condition are adaptively extracted from the S-DEM terrain feature database, a TIN is dynamically constructed, and a digital elevation model adapted to the scale is generated. Generally, the specified scale information is provided by the user as the application requirement scale.

For specific implementation, the method for constructing an adaptive scale DEM that maintains topographical features can be found in Wuhan University, Zhou Qiming, An adaptive scale DEM modeling method that maintains topographical features, Chinese patent, 201110033539, published on 2011-07-13.

For the convenience of implementation and reference, the specific steps of the embodiment are provided as follows, as shown in Figure 2:

Step 1.1, using the feature point extraction algorithm and feature line extraction algorithm for the fine-scale DEM, by changing the parameters of the feature extraction algorithm, obtain the topographic and hydrological features at different parameter levels, and construct a TIN, compare the accuracy with the original DEM, and calculate the error value. Combined with the national DEM accuracy specification, the functional relationship between the error value and the scale is established, and finally the relationship between the parameters and the scale of the hydrological model is summarized.

The embodiment first uses the existing maximum z-tolerance algorithm to extract the feature points of the ground surface from the fine-scale DEM database. The z-tolerance specifies the maximum elevation error within the tolerance range of the TIN generated by the feature point set under this level (denoted as z), with the change of z value, the feature points of different levels of the surface are retrieved from the original fine-scale DEM. In order to highlight key watershed features, the D8 algorithm is used to identify the supplementary watershed feature lines from the original fine-scale DEM and add them to the feature point set. Based on the surface feature points and watershed feature lines, finally generate a watershed-constrained TIN. Compare the original fine-scale DEM with the obtained irregular triangulation, calculate the root mean square error RMSE of the maximum elevation error z at different values through precision analysis, and perform curve fitting on the maximum elevation error z and the root mean square error RMSE , to obtain the optimal functional analytical expression between the maximum elevation error z and the root mean square error RMSE.

In step 1.2, an adaptive scale DEM data structure is constructed according to feature point sets of different scales. This step assigns scale attributes to terrain features. Various relevant specifications and standards can be used to extract feature points and lines based on scale information, so as to obtain multi-scale feature points and lines that meet the specifications. By selecting graded equidistant scales, calculating feature points and lines of specified scales, constructing an initial graded scale feature library, and extracting scale information of feature points and lines from it, assigning scale attributes to terrain features, and constructing an S-DEM feature library.

Embodiment First, according to the relationship between the scale and the contour distance in the drawing specification, obtain the value range of z and RMSE corresponding to the scale, determine the accurate value range of z, and use CPE again to extract the feature points at different scales; if in a certain scale, if there are two feature points whose spacing is smaller than the grid cell spacing of the grid digital elevation model at this scale, only the feature point with a larger value of the maximum elevation error z will be kept; if a feature point and the simplified watershed If the ratio of the line spacing to the length of the simplified watershed line is less than the set threshold, the feature point is removed.

After extracting the set of feature points conforming to the specification according to the above operations (feature lines can also be regarded as multiple feature points), assign scale attributes to each feature point, and the scale attribute of a feature point is all the scales where the feature point The coarsest scale means that the feature point is included from the finest scale to the coarsest scale; thus, a scale attribute map containing all feature points is obtained, and an adaptive scale DEM data structure is constructed. The information can be stored in the form of database in advance, that is, the mapping relationship between terrain feature points, lines and scales can be constructed, scale information attributes can be assigned to terrain features, and scale attribute maps can be obtained to construct the S-DEM terrain feature database.

Step 1.3, perform adaptive transformation at different scales. For user-defined scales, judge whether they exist in the S-DEM terrain feature database. If it exists, extract the feature points and lines at this scale from the S-DEM feature library, dynamically construct TIN, and establish a digital elevation model; if not, use the judgment criterion to obtain the user-defined scale from the features extracted at the fine scale The feature points and lines of the S-DEM terrain feature database are dynamically updated, so that the feature points and lines can still be retrieved from the S-DEM terrain feature database, and the TIN can be dynamically constructed to establish a digital elevation model;

When the user specifies the application requirement scale, if the scale specified by the user already exists in the S-DEM terrain feature database, the feature points under this scale are composed of all feature points with this scale and coarser scale attributes in the scale attribute map; If the scale specified by the user does not exist in the scale attribute map, repeat the above steps 1.1 and 1.2 to extract the feature points at this scale, and update the scale attribute map, and the feature points at this scale are determined by the updated scale attribute map All feature points with this scale and coarser scale attributes.

Step 2. Calculate the water flow direction line of each triangular surface in the irregular triangular network obtained in step 1, construct a water flow path from any position to the outlet of the basin, and establish a pipeline topology network based on the obtained water flow path.

The embodiment calculates the water flow direction of each triangle in the TIN on the basis of the adaptive scale S-DEM terrain expression, and constructs a water flow line from any position to the outlet of the watershed, thereby simplifying the three-dimensional terrain expression into a one-dimensional topological pipeline structure , establish the topological relationship between pipelines, and realize the nested multi-scale and multi-level topology structure, that is, the pipeline topology network (TFN). Referring to accompanying drawing 3, embodiment concrete steps are as follows:

In step 2.1, the irregular triangular network is divided into regular grids, and a sampling point is randomly selected in each grid as the rainfall source point.

This step implements sampling of surface water source points (such as rainfall source points). Step 1 has extracted feature points and lines from the S-DEM feature library according to the specified scale information, dynamically generated TIN subject to watershed constraints, and generated a digital elevation model. The TIN structure enables the separation of surface and water source point sampling. By comparing and analyzing different sampling methods, such as regular grid sampling, random sampling and restricted random sampling, a random sampling method based on grid restrictions is adopted, that is, the area-wide surface The TIN is divided into regular grids, and a sampling point is randomly selected in each grid as the rainfall source point. The rainfall source point is used as the starting point of pipeline tracking.

Step 2.2, by calculating the slope and aspect of each triangular surface on the irregular triangular network, the water flow direction line of each triangular surface is obtained, starting from any rainfall source point, tracing according to the water flow direction line, and constructing from any position to The water flow path at the outlet of the watershed.

This step generates the water flow direction of each triangular surface in the dynamic triangular irregular network (TIN), so as to determine the water flow direction from any point in the target area to the outlet of the watershed. By calculating the constant aspect and slope of each surface on the TIN, the three-dimensional flow direction lines of each surface are obtained. Starting from any rainfall source point in the target area, follow the direction of water flow to trace the pipeline in sequence. There are many possible flow modes for the surface water flow on adjacent triangular faces. According to these flow modes, the water flow path of each drop of rainwater from falling on the ground to the outlet of the basin can be traced, that is, the water flow path at any point can be traced.

In the embodiment, each triangular surface on the TIN constrained by the watershed has a constant slope and aspect, and the three nodes of the triangular surface can be expressed as P ₁ (x ₁ , y ₁ , z ₁ ), P ₂ (x ₂ , y ₂ , z ₂ ) and P ₃ (x ₃ , y ₃ , z ₃ ), the entire triangle is expressed as

z=f(x,y)=ax+by+c

$\left.\begin{array}{c}\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}\\ \mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}\\ \mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; b</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\end{array}\right\}$

Therefore, the slope and aspect of each triangle can be expressed as

Through the constant slope and aspect of each triangular surface on the TIN, the water flow direction line PQ representing the water flow direction of the triangular surface is obtained, where the three-dimensional coordinates of point P are (x _P , y _P , z _P ), and the three-dimensional coordinates of point Q are (x _Q ,y _Q ,z _Q ). The direction of PQ represents the slope aspect of the surface, and the length represents the slope of the surface. The coordinates of point Q are calculated from the coordinates of point P, slope (β) and aspect (α):

$\left\{\begin{array}{c}\frac{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}\\ {\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; )</span>\\ {\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; )</span>\\ {\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; )</span>\end{array}\right.$

Among them, f ₁ (.), f ₂ (.) and f ₃ (.) represent corresponding calculation functions, and the specific calculation implementation belongs to the prior art.

There are three possible flow modes for surface water flow on adjacent triangular faces:

1. If the flow direction of adjacent triangular faces deviates from the common side, the water flow path passes through the adjacent triangular faces;

2. If the flow direction of adjacent triangular faces points to the common side, it means that the surface water flow reaches a V-shaped valley, and the water flow path flows along the common side to the downstream node;

3. If the common edge of the V-shaped valley terminates (neither has a connected downstream V-shaped edge), the water flows towards the triangular face with the steepest downstream slope. If no such triangle is found, the flow path ends.

After calculating the water flow direction line of each triangular surface, the water flow path of each drop of rainwater from falling to the ground to the outlet of the watershed can be traced according to the above water flow method. A water flow path is formed by connecting water flow line segments on multiple triangular surfaces. If the adjacent triangular faces or the common edge are flat, that is, the slope of the triangular face is zero, then the flow direction of the triangular face cannot be determined. In order to eliminate the ambiguous water flow direction in the flat area, a repeated depth search method can be used to search the triangular faces in the area until the water flow direction can be determined.

In step 2.3, a pipeline topology network is established based on the water flow path obtained in step 2.2.

Embodiment Based on the three-dimensional water flow path obtained in step 2.2, analyze the topological relationship between point-point, point-line and line-line in each water flow path, and establish and store point-point, point-line and The topological relationship between lines and lines is used to construct a one-dimensional pipeline topology network. The topology network contains two tables, one stores node information, such as node ID, X, Y, Z coordinates, and other hydrological parameters, such as vegetation type of rainfall source point, soil moisture and other information; the other stores node- Line segment information, such as the beginning and end node ID, various hydrological parameters, such as the slope, length, and flow velocity of the water flow line segment. Wherein the flow velocity of the water flow line section can be deduced by some classical hydrological formulas, and the Manning formula is adopted in this embodiment, that is, v=R ^2/3 .S ^1/2 /n, wherein v represents the flow velocity (m/s), and R represents The hydrological radius (m) can be estimated by the water flow depth, S represents the slope, and n represents the Manning resistance coefficient.

In the embodiment, the topological relationship between each water flow line segment can be automatically generated from the table storing line segment information. Since each water flow line segment stores the start and end node IDs, and two adjacent line segments must share the same node, the upstream water flow line segment The end node ID of is the start node ID of the downstream water flow line segment. For example, the end node ID (616) of the line segment (213) is the same as the start node ID (616) of the line segment (214), which means that the line segment (214) is located downstream of the line segment (213). Similarly, the upstream and downstream topological relationships of all water flow line segments can be extracted, and the line segment ID and the adjacent downstream line segment IDs can be stored in the table.

Step 3, based on the pipeline topology network obtained in step 2, perform dynamic simulation of surface water flow. According to the simulated rainfall data and/or the rainfall data in the real environment and various environmental parameters, hydrological formulas can be introduced to calculate the flow velocity and perform water flow dynamic simulation. In specific implementation, the dynamic simulation of surface water flow can be carried out under the condition that only the gravity factor is considered. Simulations of surface water flow dynamics can also be performed taking gravity and other environmental variables into account. The simulation results obtained by these two methods are of technical significance, and one or both of them can be selected according to the needs of users during specific implementation. For the latter method, a variety of existing hydrological models can be used to simulate runoff for rainfall data in real environments, and then the runoff data of each model can be used to simulate water flow dynamics based on TFN. Referring to accompanying drawing 4, embodiment adopts two ways successively, concrete steps are as follows:

Step 3.1, for the simulated rainfall process (computer simulation, non-real rainfall data), simulate the rainfall process with uniform and non-uniform distribution in time and space for a period of time, only considering the gravity factor, and not considering other hydrological processes such as infiltration and evaporation for the time being In the case of , according to the pipeline topology network obtained in the above step 2, the classical hydrological formula is selected to calculate the flow velocity of each line segment, so as to obtain the rainfall-runoff simulation curve at any point in the rainfall area at any time period.

In the embodiment, a DEM with a resolution of 1091×892 and a resolution of 5m is selected as the experimental area, and a 12mm rainfall from southeast to northwest within a period of 20 minutes is simulated. Through temporal and spatial interpolation, random sampling of rainfall data under grid constraints is obtained. Since only gravity is considered and infiltration, evaporation, etc. are not considered, all rainfall is used for runoff production. According to the above steps, the pipeline topology network is established, and the flow velocity of each water flow line segment is calculated according to the Manning formula, and a series of time intervals (such as t=9s, 127s, 402s, 734s, 938s, 1120s) are selected to calculate the real-time flow rate in each water flow path. Flow, and three locations in the upper, middle and lower reaches were selected for detection, and the rainfall-runoff simulation curve of each point in the time period was generated.

Step 3.2, using real rainfall data, not only considering gravity factors, but also comprehensively considering various hydrological processes such as interception, infiltration, evaporation (adding environmental and meteorological factors, such as land cover and use type, leaf area index, air temperature and surface temperature, etc. ), using existing hydrological models (such as SWAT, SHE) to calculate runoff data, and then dynamically simulate the confluence results at different times according to the pipeline topology network.

In the embodiment, the Jinghe River Basin is selected as the experimental area in the real environment, a 3872×4937 DEM with a resolution of 90m is taken, and the pipeline topology network is generated according to the above steps. According to the daily rainfall data in 2005, environmental and meteorological factors (land cover and use type, leaf area index, air temperature and surface temperature, etc.), hydrological models such as SWAT and SHE are used to calculate the runoff data first, and then use the runoff data and the assembly line The topological network simulates the confluence results of the locations of the two main flow monitoring stations at different times. Similarly, the rainfall-runoff simulation curve at any point in the rainfall area at any time period can be obtained.

In order to verify the dynamic simulation accuracy of the pipeline topology network at multiple scales, the embodiment of the present invention finally evaluates the simulation results at different scales. The simulation results here refer to the simulation results obtained by using the hydrological model considering the gravity factor and other environmental variables. Referring to accompanying drawing 5, embodiment concrete steps are as follows:

Step a, for the existing hydrological models (such as SWAT, SHE), it is necessary to resample the fine-scale DEM into a variety of coarse-scale DEMs for multi-scale simulation, comparison and analysis. Combined with environmental and meteorological parameters at different scales, the hydrological model is used to calculate the runoff data, and then the pipeline topology network (TFN) obtained in the present invention, the existing hydrological model (such as SWAT, SHE, and step 3.2) are used to calculate the runoff data Consistent with the hydrological model) to simulate the confluence results.

In the embodiment, the original 90m resolution DEM needs to be resampled, and a series of scales are selected, for example, the resolutions are 150m, 250m, 500m, 1000m and so on. Combined with rainfall data, environmental and meteorological factors at different scales, the SWAT and SHE models are used to calculate runoff and confluence data. Extract the corresponding feature points and lines from the self-adaptive scale DEM database (S-DEM) according to different scales, construct the TIN constrained by the watershed at this scale, and generate the pipeline topology network according to step 2. Then, according to the runoff data obtained by each hydrological model, the pipeline topology network and hydrological model were used to simulate the confluence results respectively, so as to compare the simulation results of water flow at different scales.

In step b, the commonly used hydrological statistical parameters are selected as evaluation indicators for water flow simulation, such as the Nash efficiency coefficient E, which is used to evaluate the predictive ability of the hydrological model; the correlation coefficient R, which is used to measure the correlation between the measured flow rate and the simulated flow rate; the balance coefficient B , which is used to represent the ratio of the measured flow rate to the simulated flow rate. The closer these three coefficients are to 1, the more accurate the simulation results are. The hydrological statistical parameter index is calculated according to the simulation result of step a and the hydrological actual measurement data evaluation, and the specific calculation belongs to the prior art. The accuracy and consistency of various hydrological methods for simulating surface water flow at different scales are obtained through comparative testing.

The above description is only an embodiment of the present invention, and is not intended to limit the present invention. For example, the simulation results of step 3 of the present invention are directly used without evaluation. All modifications, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (3)

1. A pipeline topology network dynamic simulation method based on adaptive scale DEM, is characterized in that, comprises the following steps: Step 1, based on the self-adaptive scale DEM data structure, generate the corresponding irregular triangular network according to the specified scale information; Step 2, calculating the water flow direction line of each triangular surface in the irregular triangular network obtained in step 1, constructing a water flow path from any position to the outlet of the basin, and establishing a pipeline topology network according to the obtained water flow path; the implementation method includes the following sub-steps, Step 2.1, divide the irregular triangular network into regular grids, and randomly select a sampling point in each grid as the rainfall source point; Step 2.2, by calculating the slope and aspect of each triangular surface on the irregular triangular network, the water flow direction line of each triangular surface is obtained, starting from any rainfall source point, tracing according to the water flow direction line, and constructing from any position to the water flow path at the outlet of the watershed; Step 2.3, based on the water flow path obtained in step 2.2, establish a pipeline topology network; Step 3, based on the pipeline topology network obtained in step 2, perform dynamic simulation of surface water flow. 2. according to claim 1, the pipeline topology network dynamic simulation method based on self-adaptive scale DEM is characterized in that: in step 3, under the situation that only considers gravity factor, carry out surface water flow dynamic simulation, the realization mode is as follows, simulates a period of time For the rainfall process of uniform and non-uniform distribution in internal space and time, according to the pipeline topology network obtained in step 2, the flow velocity of each water streamline segment in the pipeline topology network is calculated using hydrological formulas, and the rainfall-runoff of any point in the rainfall area at any time period is obtained simulated curves. 3. according to the described method for dynamic simulation of pipeline topology network based on self-adaptive scale DEM of claim 1, it is characterized in that: in step 3, under the situation of considering gravity factor and other environmental variables, carry out dynamic simulation of surface water flow, the implementation is as follows, Using the real rainfall data, the hydrological model is used to calculate the runoff data, and according to the pipeline topology network obtained in step 2, the confluence results at different times are dynamically simulated.

Images