Basin-river-reservoir system coupling simulation method, device and medium

Abstract

The present invention discloses a method, device and medium for simulating a coupled watershed-river-reservoir system. The method comprises the following steps: acquiring data, establishing a watershed network diagram based on digital topographic and geomorphological information of the watershed, establishing a dynamic hydrological response unit in combination with the spatiotemporal dynamic changes of the underlying surface, and constructing a watershed water cycle simulation model based on time-varying gain; simulating the storage and regulation effects of a water conservancy project group on the runoff process of rivers and reservoirs in the watershed based on the differences in tasks of different scales of water conservancy project groups, and providing real-time feedback on the natural water cycle simulation; establishing a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model based on underwater topographic data and the required precision of river and reservoir water network simulation, and establishing a one- and two-dimensional hydrodynamic model coupler based on physical mechanism-deep learning bidirectional drive; utilizing the watershed water cycle model to drive the river and reservoir water network hydrodynamic model to couple with online reservoirs, pumping stations, and culvert projects, and determining the hydrodynamic effects of the river and reservoir water network under the operation and scheduling of the water conservancy project group under set water inflow conditions, so as to improve the calculation efficiency of hydrological and hydrodynamic simulation.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrological and hydrodynamic simulation, and in particular to a method, device and medium for simulating the coupling of a watershed and a river reservoir system. Background Art

[0002] Under the influence of high-intensity human activities, the water cycle in the basin is affected by a group of water conservancy projects including reservoirs, pumping stations, culverts, etc., forming a complex basin-river-reservoir system. Its hydrological and hydrodynamic processes are not only directly related to the natural precipitation-runoff-confluence process affected by the spatial heterogeneity of the underlying surface and the topography and landforms above and below the water, but the dynamic process of the river-reservoir water network is also affected by a series of factors such as the flattening and displacement of the reservoir, the opening and closing of pumping stations for pumping and drainage, and the diversion of culverts for flood control. How to effectively simulate the movement process of the basin-river-reservoir water network under the influence of a complex group of water conservancy projects is the key to achieving flood and drought disaster prevention and water resources scheduling.

[0003] According to a literature review, watershed hydrological models are effective methods for analyzing the natural water cycle, including precipitation, runoff, evaporation, soil water, and runoff from overslope rivers. These models typically generalize the watershed as a whole or divide it into more complex computational units based on the spatial diversity of the underlying surface to represent the vertical transformation of five waters: atmospheric water, plant water, surface water, soil water, and groundwater. Regarding the lateral flow, conceptual or kinematic wave models are used to represent overslope runoff and runoff from overslope rivers. However, these models often lack detailed representation and fail to capture the hydrodynamic processes of river-reservoir networks. Hydrodynamic models are primarily used to simulate the flow dynamics of river networks, but lack basin-level runoff simulations. Runoff models are typically used as boundary condition inputs, and basin-wide hydrodynamic simulations are often computationally intensive, making efficient computation difficult. In particular, both traditional hydrological and hydrodynamic models fail to adequately account for the complex impacts of hydraulic engineering projects, making it difficult to reflect their interplay with the natural water cycle.

[0004] Based on this, it is still necessary to further consider the deep integration of the vertical and horizontal directions of the water cycle from the perspective of the watershed-river-reservoir system, and at the same time fully consider the interaction and feedback relationship between the water conservancy project group and the natural water cycle, and further improve the design of the water cycle simulation algorithm of the watershed-river-reservoir system. We start from the watershed water cycle simulation framework of time-varying gain, consider its interaction with the river-reservoir system, and construct a coupled simulation method for the watershed-river-reservoir system. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the prior art, the object of the present invention is to provide a method, device and medium for simulating the coupling of a watershed-river-reservoir system.

[0006] Therefore, the technical solution of the present invention is:

[0007] According to a first aspect of the present invention, a method for simulating a coupled watershed-river-reservoir system is provided, the method comprising:

[0008] Acquiring data, including basic geographic information of the watershed, underlying surface data, underwater topography data, hydrological and meteorological data, and water conservancy project data;

[0009] A watershed network diagram is established based on the digital topographic information of the watershed, and a dynamic hydrological response unit is established based on the spatiotemporal dynamic changes of the underlying surface. A watershed water cycle simulation model is constructed based on the time-varying gain theory.

[0010] Based on the watershed water cycle simulation model, the regulation and storage effect of the water conservancy project group on the runoff process of the river and reservoir in the basin is simulated according to the differences in the scale and tasks of the water conservancy project group, and the natural water cycle simulation is fed back in real time;

[0011] Based on underwater topographic data and the precision requirements of river and reservoir water network simulation, a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model are established. Based on the physical mechanism and deep learning bidirectional drive, a one-dimensional and two-dimensional hydrodynamic model coupler is established.

[0012] Based on the one- and two-dimensional hydrodynamic model coupler, the basin water cycle model is used to drive the river and reservoir water network hydrodynamic model, and the online reservoirs, pumping stations, and culvert projects are coupled to determine the river and reservoir water network hydrodynamic effects of the operation and scheduling of the water conservancy project group under the set water inflow conditions.

[0013] Furthermore, the basic geographic information of the basin includes digital elevation model data and digital river network data of the basin; the underlying surface data includes land use type data and soil type data; the underwater terrain data includes large-section data and underwater terrain point cloud data; the hydrological and meteorological data includes air pressure, temperature, precipitation, humidity, wind speed, sunshine data of meteorological stations in and around the basin and runoff observation data of hydrological stations; the water conservancy project data includes engineering parameter data of reservoirs, water intakes, drainage outlets, pumping stations, culverts, and flood storage areas in the basin.

[0014] Furthermore, a watershed network diagram was established based on the digital topographic information of the watershed, and a dynamic hydrological response unit was established in combination with the spatiotemporal dynamic changes of the underlying surface. A watershed water cycle simulation model was constructed based on the time-varying gain theory, specifically including:

[0015] Based on the digital elevation model of the watershed, the steepest slope method is used in combination with the digital river network geographic information data of the watershed to divide the sub-watersheds and define the river network and encode it to establish a watershed network map;

[0016] Based on the watershed network diagram, a dynamic hydrological response unit is established taking into account the historical changes in underlying land use types and the spatial distribution of soil types. The dynamic hydrological response unit can describe the dynamic characteristics of the underlying surface under the changes in different types of land use;

[0017] Based on the time-varying gain theory, precipitation, evaporation, runoff, soil water, and groundwater flow are calculated in each dynamic hydrological response unit. The calculation formula is as follows:

[0018] P i +AW i =AW i+1 +RS i +ET i +RI i +RG i (1)

[0019]

[0020] Where, P i is the rainfall in the i-th period, AW i is the soil moisture content in the i-th period, ET i is the evapotranspiration of the i-th period, RS i is the surface runoff in the i-th period, RI i is the soil flow in the i-th period, RG i is the groundwater runoff in the i-th period, i is the period number, AW is the soil moisture content, P is the rainfall, f(AW) is the function expression, AW j+1 is the soil moisture value of the j+1th iteration, AW j is the soil moisture value of the jth iteration, f(AW j ) is the jth iteration function value, f′(AW j ) is the derivative value of the jth iteration;

[0021]

[0022] Where g1, g2 are time-varying gain runoff factors, WM is the maximum soil water content, K r ,K rg are the runoff coefficients of soil flow and groundwater runoff, α is the slope, L is the slope length, WFC is the field water holding capacity, AW u,i is the upper soil moisture content in the i-th period, AW d,i is the soil moisture content in the lower layer during the i-th period, AW d,i+1 is the soil moisture content in the lower layer during the i+1th period;

[0023]

[0024] Where E is evapotranspiration, T,E i ,E sare vegetation transpiration, canopy interception evaporation, and soil evaporation, respectively; min is the minimum function; λ is the latent heat of evaporation; Δ is the gradient of the saturated water vapor pressure-temperature curve; ρ is the air density; C p is the constant pressure specific heat of air, R nc ,R ns is the net radiation of the canopy and the surface, G is the soil heat flux, γ is the dry-bulb constant, f w is the wetted area ratio, r ac is the canopy surface boundary layer impedance, r as is the soil surface boundary layer impedance, r s is the soil impedance, r c is the canopy impedance, LW is the leaf water content, and D0 is the vapor pressure difference.

[0025] Furthermore, based on the watershed water cycle simulation model, and in accordance with the differences in the scale and tasks of water conservancy project groups, the regulation and storage effects of water conservancy project groups on the runoff process of rivers and reservoirs in the watershed are simulated, and real-time feedback is provided for the natural water cycle simulation, specifically including:

[0026] For a reservoir group, consider the differences in tasks for reservoirs of different sizes:

[0027] For each sub-basin, the small and medium-sized reservoirs are integrated according to their series and parallel characteristics to construct a sub-basin equivalent reservoir. The inflow of the sub-basin equivalent reservoir is the slope flow outflow of the sub-basin. The equivalent storage and discharge model is used for reservoir regulation, and the outflow is fed back to the sub-basin outflow.

[0028] For large-scale reservoir groups with controllable characteristics, the large-scale reservoir groups are integrated into the basin network diagram. Based on the operation regulations of large-scale reservoirs, a simulation operation storage and discharge model is established to simulate the feedback effect of large-scale reservoir group operation on river network confluence, and participate in the confluence calculation of lower-level rivers in the basin network.

[0029] The pumping and drainage or flood diversion effects of the pumping station and culvert on the river network during the consideration period are simulated based on the project scale capacity. The storage and discharge model divides the reservoir operation into the pre-flood drawdown period, flood season, water storage period, and water supply period. The calculation equation is as follows:

[0030]

[0031] Where Q out,preflood ,Q out,flood ,Q out,storage ,Q out,supply are the discharge flows during the pre-flood period, flood season, water storage period and water supply period, respectively. safe ,Q cap are safety flow, discharge capacity, V av ,V min ,V max ,V controlare the adjustable water storage capacity at the current moment, the minimum allowable water storage capacity, the maximum allowable water storage capacity and the water storage capacity controlled during the water supply period, Q eco is the ecological flow, Q demand is the socioeconomic water demand, t is time, min is the minimum function, and max is the maximum function.

[0032] Furthermore, based on the underwater topography data and the simulation precision requirements of the river and reservoir network, a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model were established, specifically including:

[0033] According to the characteristics of the river network, the river and reservoir network structure is established. According to the simulation precision requirements, a one-dimensional hydrodynamic model is established for the case where the longitudinal distance is greater than the horizontal and vertical directions, and a two-dimensional hydrodynamic model in the horizontal and vertical directions is established for lakes;

[0034] The calculation formula of the one-dimensional hydrodynamic model is as follows:

[0035]

[0036] Where A is the cross-sectional area, t is the time, Q is the flow rate, x is the flow rate, q is the interval unit flow rate, g is the acceleration of gravity, B is the water surface width, Z is the water level, n is the roughness, and R is the wetted perimeter;

[0037] The calculation formula of the two-dimensional hydrodynamic model is as follows:

[0038]

[0039] Where h is the water depth, u and v are the average velocity components of the water depth in the x and y directions respectively, and τ xx ,τ yx ,τ xy ,τ yy is the viscous stress, τ zxb ,τ zyb is the shear stress of the riverbed, τ zxs ,τ zys are the free surface stress, ρ is the density, and f is the Coriolis coefficient.

[0040] Furthermore, based on the physical mechanism and deep learning bidirectional drive, a one- and two-dimensional hydrodynamic model coupler is established, specifically including:

[0041] For the one-dimensional and two-dimensional hydrodynamic models, a coupler model is established at the coupling node. The coupler model is constructed based on the physical mechanism-deep learning bidirectional driving method. The deep learning model adopts the GRU model and a gating mechanism, including a reset gate and an update gate. The following formula is used to control and process input, memorize information, and make simulations and predictions at the current time step:

[0042]

[0043] Where x t is the model input at time t, r t is the reset gate output, z t is the update gate output, is the candidate hidden state, h t ,h t-1 are the hidden states at the current moment and the previous moment respectively, W r ,W z ,W are the reset gate weight matrix, update gate weight matrix and candidate hidden state weight matrix respectively, b r ,b z , b are the reset gate bias term, update gate bias term and bias term respectively, σ is the sigmoid activation function, and tanh is the hyperbolic tangent activation function.

[0044] The model loss function uses the following equation:

[0045]

[0046] In the formula, m, M are the sequence number and total number respectively, y i , are the simulated and measured values, respectively; f(q,z) is the water level and flow rate relationship equation of the one- and two-dimensional hydrodynamic model.

[0047] According to a second aspect of the present invention, there is provided a basin-river-reservoir system coupling simulation device, the device comprising:

[0048] A data acquisition unit is configured to acquire data, wherein the data includes basic geographical information of the watershed, underlying surface data, underwater topography data, hydrological and meteorological data, and water conservancy project data;

[0049] The first model building unit is configured to establish a watershed network diagram based on the digital topographic information of the watershed, establish a dynamic hydrological response unit based on the spatiotemporal dynamic changes of the underlying surface, and build a watershed water cycle simulation model based on the time-varying gain theory;

[0050] The second model building unit is configured to simulate the regulation and storage effect of the water conservancy project group on the runoff process of the river reservoir in the basin based on the water cycle simulation model of the watershed and according to the differences in the scale and tasks of the water conservancy project group, and provide real-time feedback on the natural water cycle simulation;

[0051] The third model building unit is configured to establish a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model based on underwater topography data and the required precision of river and reservoir water network simulation, and to establish a one-dimensional and two-dimensional hydrodynamic model coupler based on a two-way drive of physical mechanism and deep learning;

[0052] The hydrodynamic effect determination unit is configured to drive the river and reservoir water network hydrodynamic model based on the one- and two-dimensional hydrodynamic model coupler using the basin water cycle model, and couple the online reservoirs, pumping stations, and culvert projects to determine the river and reservoir water network hydrodynamic effect of the operation and scheduling of the water conservancy project group under the set water inflow conditions.

[0053] Furthermore, the basic geographic information of the basin includes digital elevation model data and digital river network data of the basin; the underlying surface data includes land use type data and soil type data; the underwater terrain data includes large-section data and underwater terrain point cloud data; the hydrological and meteorological data includes air pressure, temperature, precipitation, humidity, wind speed, sunshine data of meteorological stations in and around the basin and runoff observation data of hydrological stations; the water conservancy project data includes engineering parameter data of reservoirs, water intakes, drainage outlets, pumping stations, culverts, and flood storage areas in the basin.

[0054] Furthermore, the coefficient determination module is further configured to:

[0055] Based on the digital elevation model of the watershed, the steepest slope method is used in combination with the digital river network geographic information data of the watershed to divide the sub-watersheds and define the river network and encode it to establish a watershed network map;

[0056] Based on the watershed network diagram, a dynamic hydrological response unit is established taking into account the historical changes in underlying land use types and the spatial distribution of soil types. The dynamic hydrological response unit can describe the dynamic characteristics of the underlying surface under the changes in different types of land use;

[0057] Based on the time-varying gain theory, precipitation, evaporation, runoff, soil water, and groundwater flow are calculated in each dynamic hydrological response unit. The calculation formula is as follows:

[0058] P i +AW i =AW i+1 +RS i +ET i +RI i +RG i (1)

[0059]

[0060] Where, P i is the rainfall in the i-th period, AW i is the soil moisture content in the i-th period, ET i is the evapotranspiration of the i-th period, RS i is the surface runoff in the i-th period, RI i is the soil flow in the i-th period, RG iis the groundwater runoff in the i-th period, i is the period number, AW is the soil moisture content, P is the rainfall, f(AW) is the function expression, AW j+1 is the soil moisture value of the j+1th iteration, AW j is the soil moisture value of the jth iteration, f(AW j ) is the jth iteration function value, f′(AW j ) is the derivative value of the jth iteration;

[0061]

[0062] Where g1, g2 are time-varying gain runoff factors, WM is the maximum soil water content, K r ,K rg are the runoff coefficients of soil flow and groundwater runoff, α is the slope, L is the slope length, WFC is the field water holding capacity, AW u,i is the upper soil moisture content in the i-th period, AW d,i is the soil moisture content in the lower layer during the i-th period, AW d,i+1 is the soil moisture content in the lower layer during the i+1th period;

[0063]

[0064] Where E is evapotranspiration, T,E i ,E s are vegetation transpiration, canopy interception evaporation, and soil evaporation, respectively; min is the minimum function; λ is the latent heat of evaporation; Δ is the gradient of the saturated water vapor pressure-temperature curve; ρ is the air density; C p is the constant pressure specific heat of air, R nc ,R ns is the net radiation of the canopy and the surface, G is the soil heat flux, γ is the dry-bulb constant, f w is the wetted area ratio, r ac is the canopy surface boundary layer impedance, r as is the soil surface boundary layer impedance, r s is the soil impedance, r c is the canopy impedance, LW is the leaf water content, and D0 is the vapor pressure difference.

[0065] According to a third aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, executes the watershed-river-reservoir system coupling simulation method as described above.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] 1. The present invention fully considers the coupled simulation of the dynamic process of the basin-river network under the joint action of water conservancy project groups including reservoirs, pumping stations, culverts, etc., and simulates the impact of water conservancy project groups on runoff generation and convergence through storage and discharge models, thereby improving the accuracy of runoff simulation.

[0068] 2. The present invention combines the advantages of the basin hydrological model and the hydrodynamic model, and fully integrates the physical mechanism and artificial intelligence simulation method to construct a model coupler, thereby improving the calculation efficiency of hydrological and hydrodynamic simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In the drawings, which are not necessarily drawn to scale, the same reference numerals may describe similar components in different views. The same reference numerals with letter suffixes or different letter suffixes may represent different instances of similar components. The accompanying drawings generally illustrate various embodiments by way of example and not limitation, and together with the description and claims, serve to illustrate the embodiments of the invention. Where appropriate, the same reference numerals are used throughout the drawings to refer to the same or similar parts. Such embodiments are illustrative and are not intended to be exhaustive or exclusive of the embodiments of the present apparatus or method.

[0070] Figure 1 Flowchart of a method for simulating a watershed-river-reservoir system coupling according to an embodiment of the present invention.

[0071] Figure 2 2 is a Yangtze River Basin-River Reservoir Network Diagram according to an embodiment of the present invention.

[0072] Figure 3 2. This is a range diagram of the hydrodynamic model of the Yangtze River mainstream and the two lakes according to an embodiment of the present invention.

[0073] Figure 4 2 is a diagram showing the results of hydrological and hydrodynamic simulation of the Yangtze River Basin and river reservoirs according to an embodiment of the present invention.

[0074] Figure 5 This is a diagram showing the hydrodynamic effect of the reservoir-pump station-culvert sluice joint dispatching according to an embodiment of the present invention.

[0075] Figure 6 2 is a structural diagram of a basin-river-reservoir system coupling simulation device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0076] The following examples are merely intended to better illustrate the present invention, but the present invention is not limited to the examples set forth herein. Therefore, non-essential improvements and adjustments to the embodiments made by those skilled in the art based on the above-described invention and applied to other embodiments are still within the scope of the present invention.

[0077] The present invention will now be further described with reference to the accompanying drawings.

[0078] The embodiment of the present invention provides a method for simulating the coupling of a watershed and a river reservoir system. The method has a simple process as follows: Figure 1 As shown, the simulation method includes the following five steps.

[0079] Step 1: Data preparation: collect and organize basic geographic information of the basin, underlying surface data, underwater topography data, hydrological and meteorological data, water conservancy project data, etc.

[0080] The basic geographic information collected and organized in the basin includes digital elevation model data (DEM) and digital river network data of the basin; the underlying surface data includes land use type data and soil type data; the underwater topography data includes large-section data and underwater topography point cloud data; the hydrological and meteorological data include air pressure, temperature, precipitation, humidity, wind speed, sunshine and other data of meteorological stations in and around the basin and runoff observation data of hydrological stations; the water conservancy project data include engineering parameter data of large, medium and small reservoirs, water intakes, outlets, pumping stations, culverts, flood storage areas and other engineering parameters in the basin.

[0081] For example, this embodiment takes the Yangtze River Basin as the research object, and collects data including 30m precision DEM of the Yangtze River Basin, 1km resolution land use data and 1km soil type data of the Yangtze River Basin in 1980, 1990, 1995, 2000, 2005, 2010, 2015 and 2020; daily data of air pressure, temperature, precipitation, humidity, wind speed, sunshine, etc. of 249 meteorological stations in the Yangtze River Basin since 1980; main control stations of the Yangtze River Basin, including Zhutuo, Beibei, Cuntan, Wulong, Qingxichang, Wanxian, Miaohe, Yichang, Zhicheng, Xinjiang, etc. The data includes daily runoff data for stations such as Hukou, Shadaoguan, Mituo Temple, Kangjiagang, Guanjiapu, Shimen, Taoyuan, Taojiang, Xiangtan, Chenglingji, Luoshan, Xiantao, Hankou, Jiujiang, Hukou, Qiujin, Wanjiabu, Waizhou, Lijiadu, Meigang, Hushan, Dufengkeng, and Datong. Water conservancy project data include characteristic storage capacity and characteristic water levels of large, medium, and small reservoirs in the Yangtze River Basin, design flow of large pumping stations and sluice stations in the middle and lower reaches of the Yangtze River, embankment scope, and volume of flood storage areas. Underwater topography data include data on large sections of the Three Gorges Reservoir in the upper reaches of the Yangtze River, the middle and lower reaches of the Yangtze River, and underwater topography point cloud data of Poyang Lake and Dongting Lake.

[0082] Step 2: Establish a watershed network diagram based on the digital topographic information of the watershed, establish a dynamic hydrological response unit considering the spatiotemporal dynamic changes of the underlying surface, and construct a watershed water cycle simulation model based on the time-varying gain theory.

[0083] Based on the digital elevation model of the watershed, the steepest slope method is used in combination with the digital river network geographic information data of the watershed to divide the sub-watersheds and define the river network and encode it to establish a watershed network diagram. On this basis, considering the historical changes in the underlying land use types and the spatial distribution of soil types, a dynamic hydrological response unit is established. Unlike the traditional long-term static hydrological response unit, the dynamic hydrological response unit can more finely depict the dynamic characteristics of the underlying surface under the changes in different types of land use. Based on the time-varying gain theory, precipitation, evaporation, runoff, soil water, and groundwater flow are calculated in each hydrological response unit. The main calculation formulas are as follows:

[0084] P i +AW i =AW i+1 +RS i +ET i +RI i +RG i (1)

[0085]

[0086] Where, P i is the rainfall in the i-th period, AW i is the soil moisture content in the i-th period, ET i is the evapotranspiration of the i-th period, RS i is the surface runoff in the i-th period, RI i is the soil flow in the i-th period, RG i is the groundwater runoff in the i-th period, i is the period number, AW is the soil moisture content, P is the rainfall, f(AW) is the function expression, AW j+1 is the soil moisture value of the j+1th iteration, AW j is the soil moisture value of the jth iteration, f(AW j ) is the jth iteration function value, f′(AW j ) is the derivative value of the jth iteration;

[0087]

[0088] Where g1, g2 are time-varying gain runoff factors, WM is the maximum soil water content, K r ,K rg are the runoff coefficients of soil flow and groundwater runoff, α is the slope, L is the slope length, WFC is the field water holding capacity, AW u,i is the upper soil moisture content in the i-th period, AW d,i is the soil moisture content in the lower layer during the i-th period, AW d,i+1 is the soil moisture content in the lower layer during the i+1th period;

[0089]

[0090] Where E is evapotranspiration, T,E i ,E s are vegetation transpiration, canopy interception evaporation, and soil evaporation, respectively; min is the minimum function; λ is the latent heat of evaporation; Δ is the gradient of the saturated water vapor pressure-temperature curve; ρ is the air density; C p is the constant pressure specific heat of air, R nc ,R ns is the net radiation of the canopy and the surface, G is the soil heat flux, γ is the dry-bulb constant, f w is the wetted area ratio, r ac is the canopy surface boundary layer impedance, r as is the soil surface boundary layer impedance, r s is the soil impedance, r c is the canopy impedance, LW is the leaf water content, and D0 is the vapor pressure difference.

[0091] Basin confluence is divided into slope confluence and river confluence. Slope confluence is mainly based on the unit line method, and river confluence is mainly based on the Muskingum method. In actual calculations, based on the basin network diagram, the calculation is performed step by step from upstream to downstream until all nodes in the basin network diagram are calculated.

[0092] In this embodiment, the Yangtze River Basin is divided into 1212 sub-basins. Based on soil type data and land use type data since 1980, the sub-basins are divided into 23672 dynamic hydrological response units that are updated every five years on average. A basin network diagram is established, including 2441 network nodes and 2440 connection paths, where the nodes include sub-flows.

[0093] Step 3: Based on the basin water cycle simulation model, taking into account the differences in tasks of different scales of water conservancy project groups, simulate the regulating and storage effect of water conservancy project groups on the runoff process of rivers and reservoirs in the basin, and provide real-time feedback on the natural water cycle simulation.

[0094] To study the impact of water conservancy project clusters within a river basin on the water cycle, considering the diverse missions of reservoirs of varying scale, the small and medium-sized reservoirs within each sub-basin were integrated based on their series and parallel characteristics. A sub-basin equivalent reservoir was constructed, with the inflow to the sub-basin equivalent reservoir being the slope outflow of the sub-basin. An equivalent storage and discharge model was used for reservoir regulation, with the outflow being fed back to the sub-basin outflow. For large, controlled reservoir clusters, the clusters were integrated into the river basin network diagram, and a simulation storage and discharge model was established based on the scheduling procedures for the large reservoirs to simulate the feedback effect of the large-scale reservoir cluster operation on the river network flow. This model also incorporates the feedback effect of the large-scale reservoir operation on the river network flow calculations for the downstream rivers in the river basin network. The effects of pumping stations, culverts, and sluices on the drainage or flood diversion of river network outflows during the consideration period were simulated primarily based on the project scale capacity. The storage and discharge model divides reservoir operation into the pre-flood drawdown period, flood season, water storage period, and water supply period. The main calculation equations are as follows:

[0095]

[0096] Where Q out,preflood ,Q out,flood ,Q out,storage ,Q out,supply are the discharge flows during the pre-flood period, flood season, water storage period and water supply period, respectively. safe ,Q cap are safety flow, discharge capacity, V av ,V min ,V max ,V control are the adjustable water storage capacity at the current moment, the minimum allowable water storage capacity, the maximum allowable water storage capacity and the water storage capacity controlled during the water supply period, Q eco is the ecological flow, Q demand is the socioeconomic water demand, t is time, min is the minimum function, and max is the maximum function.

[0097] In this embodiment, 178 large control-type reservoirs were selected from more than 40,000 reservoirs of different scales and tasks in the Yangtze River Basin to directly participate in the basin network diagram simulation calculation. A simulation scheduling storage and discharge model was established based on their characteristic storage capacity, characteristic water level, discharge capacity, engineering tasks and other parameters; the remaining small and medium-sized reservoirs were integrated into 1,212 equivalent reservoirs according to their sub-basins, and equivalent storage and discharge models were constructed according to the engineering tasks for simulation calculation.

[0098] Step 4: Based on the underwater topography data and the simulation precision requirements of the river and reservoir network, establish one-dimensional and two-dimensional hydrodynamic models, construct a physical mechanism-deep learning bidirectional driving method, and establish a one-dimensional and two-dimensional hydrodynamic model coupler:

[0099] According to the characteristics of the river network, the network structure of the river and reservoir water network is established. According to the simulation precision requirements, a one-dimensional hydrodynamic model is established for the longitudinal distance much larger than the horizontal and vertical directions, and a two-dimensional hydrodynamic model is established for the horizontal and vertical directions of the lake. The main calculation formula of the one-dimensional hydrodynamic model is as follows, and the Preissmann four-point eccentric format is used for discrete solution.

[0100]

[0101] Where A is the cross-sectional area, t is time, Q is the flow rate, x is the flow rate, q is the unit flow rate of the interval, g is the acceleration of gravity, B is the water surface width, Z is the water level, n is the roughness, and R is the wetted perimeter.

[0102] The calculation formula of the two-dimensional hydrodynamic model is as follows, and it is solved using the unstructured grid finite volume method.

[0103]

[0104] Where h is the water depth, u and v are the average velocity components of the water depth in the x and y directions respectively, and τ xx ,τ yx ,τ xy ,τ yy is the viscous stress, τ zxb ,τ zyb is the shear stress of the riverbed, τ zxs ,τ zys are the free surface stress, ρ is the density, and f is the Coriolis coefficient.

[0105] For one-dimensional hydrodynamic models and two-dimensional hydrodynamic models, a coupler model is established at the coupling node. Considering the physical mechanism-deep learning bidirectional driving method, the deep learning model adopts the GRU model and a gating mechanism with a reset gate and an update gate to control and process input, memory and other information and make simulations and predictions at the current time step.

[0106]

[0107] Where x t is the model input at time t, r t is the reset gate output, z t is the update gate output, is the candidate hidden state, h t ,h t-1 are the hidden states at the current moment and the previous moment respectively, W r ,W z ,W are the reset gate weight matrix, update gate weight matrix and candidate hidden state weight matrix respectively, b r ,b z, b are the reset gate bias term, update gate bias term and bias term respectively, σ is the sigmoid activation function, and tanh is the hyperbolic tangent activation function.

[0108] The model loss function uses the following equation:

[0109]

[0110] In the formula, m, M are the sequence number and total number respectively, y i , are the simulated and measured values, respectively; f(q,z) is the water level and flow rate relationship equation of the one- and two-dimensional hydrodynamic model.

[0111] In this example, a one-dimensional hydrodynamic model of the Yangtze River mainstream from Zhutuo to Datong (1800 km) is established, including the sections from Zhutuo to the Three Gorges Dam site and from the Three Gorges Dam site to Datong, which are divided into 397 and 535 sections respectively. A two-dimensional hydrodynamic model of Poyang Lake and Dongting Lake is constructed, which is divided into 6675 and 26659 grids respectively. The one- and two-dimensional hydrodynamic model coupler includes the water exchange simulation between Dongting Lake and the Yangtze River, and the water exchange simulation between Poyang Lake and the Yangtze River. The simulation range is shown in Figure 2. Figure 3 shown.

[0112] Step 5: Based on the coupler, the basin water cycle model is used to drive the river-reservoir water network hydrodynamic model, and online reservoirs, pumping stations, culverts and other projects are coupled to analyze the hydrodynamic effects of the river-reservoir water network under the operation and scheduling of the water conservancy project group under typical water inflow conditions.

[0113] Based on the models constructed in steps 3 and 4, the river-reservoir water network hydrodynamic model established in step 4 is driven by the coupler and the inflow boundary of the basin river control point simulated in step 3 to perform basin-reservoir coupling simulation; at the same time, the impact of online reservoirs, pumping stations, culverts and other projects in the river-reservoir water network projects on the regulation, drainage, and flood diversion of the basin water cycle simulation inflow is considered, and the hydrodynamic effect of the river-reservoir water network under the operation and scheduling of the water conservancy project group under typical water inflow conditions is analyzed.

[0114] In this example, the coupling of the Yangtze River Basin-River-Reservoir system, the coupling of the basin-reservoir hydrology and hydrodynamics is based on the daily flow data of Zhutuo, Beibei, Wulong, Shimen, Taoyuan, Taojiang, Xiangtan, Xiantao, Qiujin, Wanjiabu, Waizhou, Lijiadu, Meigang, Hushan, Dufengkeng and other stations in the basin water cycle simulation considering the role of the basin reservoir group, and the one- and two-dimensional hydrodynamic model is driven by the time downscaling coupler; on the other hand, the one- and two-dimensional hydrodynamic model is coupled based on the runoff data of the basin water cycle simulation and the coupler constructed by the deep learning model. This coupled simulation calculation significantly improves the simulation calculation accuracy, with an average calculation accuracy of about 0.8, which is about 10% higher than the traditional model accuracy. At the same time, it greatly improves the calculation efficiency of the model, shortening the traditional hydrodynamic simulation calculation from 1 day to less than 2 hours. The results of the Yangtze River Basin-River Reservoir Hydrology and Hydrodynamic Simulation are shown in the figure. Figure 4 .

[0115] This example further considers the effects of online reservoirs, pumping stations, culverts and other projects on the Yangtze River mainstream, including the Three Gorges Reservoir, large-scale drainage pumping stations and more than 390 culverts along the Yangtze River mainstream. Taking the "2012.07 Yangtze River typical flood" as an example, through combined scenario analysis (scenario settings see Table 1), the effect of the operation and scheduling of water conservancy project groups on the river and reservoir water network is studied. The operation of the Three Gorges Reservoir effectively reduced the water level in the Jingjiang River section. Taking the water level in Chenglingji as an example, the maximum reduction was about 1.0m. Pumping and drainage of the pumping station group in the middle and lower reaches of the Yangtze River may cause the water level to increase. The hydrological effect results of the combined operation of reservoirs, pumping stations and culverts are shown in Figure 5 .

[0116] Table 1 Reservoir-pumping station-culvert joint operation scenario

[0117]

[0118]

[0119] The embodiment of the present invention provides a basin-river reservoir system coupling simulation device, such as Figure 6 As shown, the device includes:

[0120] The data acquisition unit 601 is configured to acquire data, including basic geographical information of the watershed, underlying surface data, underwater topography data, hydrological and meteorological data, and water conservancy project data;

[0121] The first model building unit 602 is configured to establish a watershed network diagram based on the digital topographic information of the watershed, establish a dynamic hydrological response unit based on the spatiotemporal dynamic changes of the underlying surface, and build a watershed water cycle simulation model based on the time-varying gain theory;

[0122] The second model building unit 603 is configured to simulate the regulation and storage effect of the water conservancy project group on the runoff process of the river and reservoir in the basin based on the water cycle simulation model of the watershed and according to the differences in the scale and tasks of the water conservancy project group, and provide real-time feedback on the natural water cycle simulation;

[0123] The third model building unit 604 is configured to build a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model based on the underwater topography data and the required precision of the river and reservoir water network simulation, and to build a one-dimensional and two-dimensional hydrodynamic model coupler based on the physical mechanism and deep learning bidirectional drive;

[0124] The hydrodynamic effect determination unit 605 is configured to drive the river and reservoir water network hydrodynamic model based on the one- and two-dimensional hydrodynamic model coupler using the basin water cycle model, and couple the online reservoirs, pumping stations, and culvert projects to determine the river and reservoir water network hydrodynamic effect of the water conservancy project group operation and scheduling under the set water inflow conditions.

[0125] In some embodiments, the basic geographic information of the watershed includes digital elevation model data and digital river network data of the watershed; the underlying surface data includes land use type data and soil type data; the underwater terrain data includes large-section data and underwater terrain point cloud data; the hydrological and meteorological data includes air pressure, temperature, precipitation, humidity, wind speed, sunshine data of meteorological stations in and around the watershed and runoff observation data of hydrological stations; the water conservancy project data includes engineering parameter data of reservoirs, water intakes, drainage outlets, pumping stations, culverts, and flood storage areas in the watershed.

[0126] In some embodiments, the coefficient determination module is further configured to:

[0127] Based on the digital elevation model of the watershed, the steepest slope method is used in combination with the digital river network geographic information data of the watershed to divide the sub-watersheds and define the river network and encode it to establish a watershed network map;

[0128] Based on the watershed network diagram, a dynamic hydrological response unit is established taking into account the historical changes in underlying land use types and the spatial distribution of soil types. The dynamic hydrological response unit can describe the dynamic characteristics of the underlying surface under the changes in different types of land use;

[0129] Based on the time-varying gain theory, precipitation, evaporation, runoff, soil water, and groundwater flow are calculated in each dynamic hydrological response unit. The calculation formula is as follows:

[0130] P i +AW i =AW i+1 +RS i +ET i +RI i +RG i (1)

[0131]

[0132] Where, P i is the rainfall in the i-th period, AW i is the soil moisture content in the i-th period, ET i is the evapotranspiration of the i-th period, RS i is the surface runoff in the i-th period, RI i is the soil flow in the i-th period, RG i is the groundwater runoff in the i-th period, i is the period number, AW is the soil moisture content, P is the rainfall, f(AW) is the function expression, AW j+1 is the soil moisture value of the j+1th iteration, AW j is the soil moisture value of the jth iteration, f(AW j ) is the jth iteration function value, f′(AW j ) is the derivative value of the jth iteration;

[0133]

[0134] Where g1, g2 are time-varying gain runoff factors, WM is the maximum soil water content, K r ,K rg are the runoff coefficients of soil flow and groundwater runoff, α is the slope, L is the slope length, WFC is the field water holding capacity, AW u,i is the upper soil moisture content in the i-th period, AW d,i is the soil moisture content in the lower layer during the i-th period, AW d,i+1 is the soil moisture content in the lower layer during the i+1th period;

[0135]

[0136] Where E is evapotranspiration, T,E i ,E s are vegetation transpiration, canopy interception evaporation, and soil evaporation, respectively; min is the minimum function; λ is the latent heat of evaporation; Δ is the gradient of the saturated water vapor pressure-temperature curve; ρ is the air density; C p is the constant pressure specific heat of air, R nc ,R ns is the net radiation of the canopy and the surface, G is the soil heat flux, γ is the dry-bulb constant, f w is the wetted area ratio, r ac is the canopy surface boundary layer impedance, r as is the soil surface boundary layer impedance, r s is the soil impedance, r c is the canopy impedance, LW is the leaf water content, and D0 is the vapor pressure difference.

[0137] It should be noted that the various device structures described in this embodiment belong to the same technical concept as the method described previously, and achieve the same technical effects through the same principles, which will not be repeated here.

[0138] An embodiment of the present invention also provides a non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, executes the basin-river-reservoir system coupling simulation method as described in any of the above embodiments.

[0139] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. A method for simulating the coupled watershed-river-reservoir system, characterized in that: The method comprises: Acquiring data, including basic geographic information of the watershed, underlying surface data, underwater topography data, hydrological and meteorological data, and water conservancy project data; A watershed network diagram is established based on the digital topographic information of the watershed, and a dynamic hydrological response unit is established based on the spatiotemporal dynamic changes of the underlying surface. A watershed water cycle simulation model is constructed based on the time-varying gain theory. Based on the watershed water cycle simulation model, the regulation and storage effect of the water conservancy project group on the runoff process of the river and reservoir in the basin is simulated according to the differences in the scale and tasks of the water conservancy project group, and the natural water cycle simulation is fed back in real time; Based on underwater topographic data and the precision requirements of river and reservoir water network simulation, a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model are established. Based on the physical mechanism and deep learning bidirectional drive, a one-dimensional and two-dimensional hydrodynamic model coupler is established. Based on the one- and two-dimensional hydrodynamic model coupler, the basin water cycle model is used to drive the river-reservoir water network hydrodynamic model, and coupled with online reservoirs, pumping stations, and culvert projects to determine the river-reservoir water network hydrodynamic effects of the water conservancy project group operation scheduling under set water inflow conditions; Based on the basin water cycle simulation model, and taking into account the differences in the scale and tasks of water conservancy project groups, the regulation and storage effects of water conservancy project groups on the runoff process of rivers and reservoirs in the basin are simulated, and real-time feedback is provided for natural water cycle simulation, specifically including: For a reservoir group, consider the differences in tasks for reservoirs of different sizes: For each sub-basin, the small and medium-sized reservoirs are integrated according to their series and parallel characteristics to construct a sub-basin equivalent reservoir. The inflow of the sub-basin equivalent reservoir is the slope flow outflow of the sub-basin. The equivalent storage and discharge model is used for reservoir regulation, and the outflow is fed back to the sub-basin outflow. For large-scale reservoir groups with controllable characteristics, the large-scale reservoir groups are integrated into the basin network diagram. Based on the operation regulations of large-scale reservoirs, a simulation operation storage and discharge model is established to simulate the feedback effect of large-scale reservoir group operation on river network confluence, and participate in the confluence calculation of lower-level rivers in the basin network. The pumping and drainage or flood diversion effects of the pumping station and culvert on the river network during the consideration period are simulated based on the project scale capacity. The storage and discharge model divides the reservoir operation into the pre-flood drawdown period, flood season, water storage period, and water supply period. The calculation equation is as follows: Where Q out,preflood ,Q out,flood ,Q out,storage ,Q out,supply are the discharge flows during the pre-flood period, flood season, water storage period and water supply period, respectively. safe ,Q cap are safety flow, discharge capacity, V av ,V min ,V max ,V control are the adjustable water storage capacity at the current moment, the minimum allowable water storage capacity, the maximum allowable water storage capacity and the water storage capacity controlled during the water supply period, Q eco is the ecological flow, Q demand is the socioeconomic water demand, t is time, min is the minimum function, and max is the maximum function; Based on the underwater topography data and the simulation precision requirements of the river and reservoir network, a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model are established, including: According to the characteristics of the river network, the river and reservoir network structure is established. According to the simulation precision requirements, a one-dimensional hydrodynamic model is established for the case where the longitudinal distance is greater than the horizontal and vertical directions, and a two-dimensional hydrodynamic model in the horizontal and vertical directions is established for lakes; The calculation formula of the one-dimensional hydrodynamic model is as follows: Where A is the cross-sectional area, t is the time, Q is the flow rate, x is the flow rate, q is the interval unit flow rate, g is the acceleration of gravity, B is the water surface width, Z is the water level, n is the roughness, and R is the wetted perimeter; The calculation formula of the two-dimensional hydrodynamic model is as follows: Where h is the water depth, u and v are the average velocity components of the water depth in the x and y directions respectively, and τ xx ,τ yx ,τ xy ,τ yy is the viscous stress, τ zxb ,τ zyb is the shear stress of the riverbed, τ zxs ,τ zys are the free surface stress, ρ is the density, and f is the Coriolis coefficient.

2. The method according to claim 1, characterized in that The basic geographic information of the basin includes digital elevation model data and digital river network data of the basin; the underlying surface data includes land use type data and soil type data; the underwater terrain data includes large-section data and underwater terrain point cloud data; the hydrological and meteorological data includes air pressure, temperature, precipitation, humidity, wind speed, sunshine data of meteorological stations in and around the basin and runoff observation data of hydrological stations; the water conservancy project data includes engineering parameter data of reservoirs, water intakes, drainage outlets, pumping stations, culverts, and flood storage areas in the basin.

3. The method according to claim 1, characterized in that A watershed network diagram is established based on the digital topographic information of the watershed. A dynamic hydrological response unit is established based on the spatiotemporal dynamic changes of the underlying surface. A watershed water cycle simulation model is constructed based on the time-varying gain theory. Specifically, the following are included: Based on the digital elevation model of the watershed, the steepest slope method is used in combination with the digital river network geographic information data of the watershed to divide the sub-watersheds and define the river network and encode it to establish a watershed network map; Based on the watershed network diagram, a dynamic hydrological response unit is established taking into account the historical changes in underlying land use types and the spatial distribution of soil types. The dynamic hydrological response unit can describe the dynamic characteristics of the underlying surface under the changes in different types of land use; Based on the time-varying gain theory, precipitation, evaporation, runoff, soil water, and groundwater flow are calculated in each dynamic hydrological response unit. The calculation formula is as follows: P i +AW i =AW i+1 +RS i +ET i +RI i +RG i (1) Where, P i is the rainfall in the i-th period, AW i is the soil moisture content in the i-th period, ET i is the evapotranspiration of the i-th period, RS i is the surface runoff in the i-th period, RI i is the soil flow in the i-th period, RG i is the groundwater runoff in the i-th period, i is the period number, AW is the soil moisture content, P is the rainfall, f(AW) is the function expression, AW j+1 is the soil moisture value of the j+1th iteration, AW j is the soil moisture value of the jth iteration, f(AW j ) is the jth iteration function value, f′(AW j ) is the derivative value of the jth iteration; Where g1, g2 are time-varying gain runoff factors, WM is the maximum soil water content, K r ,K rg are the runoff coefficients of soil flow and groundwater runoff, α is the slope, L is the slope length, WFC is the field water holding capacity, AW u,i is the upper soil moisture content in the i-th period, AW d,i is the soil moisture content in the lower layer during the i-th period, AW d,i+1 is the soil moisture content in the lower layer during the i+1th period; Where E is evapotranspiration, T,E i ,E s are vegetation transpiration, canopy interception evaporation, and soil evaporation, respectively; min is the minimum function; λ is the latent heat of evaporation; Δ is the gradient of the saturated water vapor pressure-temperature curve; ρ is the air density; C p is the constant pressure specific heat of air, R nc ,R ns is the net radiation of the canopy and the surface, G is the soil heat flux, γ is the dry-bulb constant, f w is the wetted area ratio, r ac is the canopy surface boundary layer impedance, r as is the soil surface boundary layer impedance, r s is the soil impedance, r c is the canopy impedance, LW is the leaf water content, and D0 is the vapor pressure difference.

4. The method according to claim 1, wherein Based on the physical mechanism and deep learning bidirectional drive, a one- and two-dimensional hydrodynamic model coupler is established, specifically including: For one-dimensional and two-dimensional hydrodynamic models, coupler models are established at the coupling nodes. A physical mechanism-based deep learning bidirectional driving method is used to construct a coupler model. The deep learning model adopts the GRU model and a gating mechanism, including a reset gate and an update gate. The following formula is used to control and process input, memorize information, and make simulations and predictions at the current time step: Where x t is the model input at time t, r t is the reset gate output, z t is the update gate output, h% t is the candidate hidden state, h t ,h t-1 are the hidden states at the current moment and the previous moment respectively, W r ,W z ,W are the reset gate weight matrix, update gate weight matrix and candidate hidden state weight matrix respectively, b r ,b z ,b are reset gate bias term, update gate bias term and bias term respectively, σ is sigmoid activation function, tanh is hyperbolic tangent activation function; The model loss function uses the following equation: In the formula, m and M are the sequence number and total number respectively. are the simulated and measured values, respectively; f(q,z) is the water level and flow rate relationship equation of the one- and two-dimensional hydrodynamic model.

5. A basin-river-reservoir system coupling simulation device, characterized in that: The device comprises: A data acquisition unit is configured to acquire data, wherein the data includes basic geographical information of the watershed, underlying surface data, underwater topography data, hydrological and meteorological data, and water conservancy project data; The first model building unit is configured to establish a watershed network diagram based on the digital topographic information of the watershed, establish a dynamic hydrological response unit based on the spatiotemporal dynamic changes of the underlying surface, and build a watershed water cycle simulation model based on the time-varying gain theory; The second model building unit is configured to simulate the regulation and storage effect of the water conservancy project group on the runoff process of the river reservoir in the basin based on the water cycle simulation model of the watershed and according to the differences in the scale and tasks of the water conservancy project group, and provide real-time feedback on the natural water cycle simulation; The third model building unit is configured to establish a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model based on underwater topography data and the required precision of river and reservoir water network simulation, and to establish a one-dimensional and two-dimensional hydrodynamic model coupler based on a two-way drive of physical mechanism and deep learning; The hydrodynamic effect determination unit is configured to drive the river-reservoir water network hydrodynamic model using the watershed water cycle model based on the one- and two-dimensional hydrodynamic model coupler, and couple the online reservoirs, pumping stations, and culvert projects to determine the river-reservoir water network hydrodynamic effect of the water conservancy project group operation scheduling under the set water inflow condition; Based on the basin water cycle simulation model, and taking into account the differences in the scale and tasks of water conservancy project groups, the regulation and storage effects of water conservancy project groups on the runoff process of rivers and reservoirs in the basin are simulated, and real-time feedback is provided for natural water cycle simulation, specifically including: For a reservoir group, consider the differences in tasks for reservoirs of different sizes: For each sub-basin, the small and medium-sized reservoirs are integrated according to their series and parallel characteristics to construct a sub-basin equivalent reservoir. The inflow of the sub-basin equivalent reservoir is the slope flow outflow of the sub-basin. The equivalent storage and discharge model is used for reservoir regulation, and the outflow is fed back to the sub-basin outflow. For large-scale reservoir groups with controllable characteristics, the large-scale reservoir groups are integrated into the basin network diagram. Based on the operation regulations of large-scale reservoirs, a simulation operation storage and discharge model is established to simulate the feedback effect of large-scale reservoir group operation on river network confluence, and participate in the confluence calculation of lower-level rivers in the basin network. The pumping and drainage or flood diversion effects of the pumping station and culvert on the river network during the consideration period are simulated based on the project scale capacity. The storage and discharge model divides the reservoir operation into the pre-flood drawdown period, flood season, water storage period, and water supply period. The calculation equation is as follows: Where Q out,preflood ,Q out,flood ,Q out,storage ,Q out,supply are the discharge flows during the pre-flood period, flood season, water storage period and water supply period, respectively. safe ,Q cap are safety flow, discharge capacity, V av ,V min ,V max ,V control are the adjustable water storage capacity at the current moment, the minimum allowable water storage capacity, the maximum allowable water storage capacity and the water storage capacity controlled during the water supply period, Q eco is the ecological flow, Q demand is the socioeconomic water demand, t is time, min is the minimum function, and max is the maximum function; Based on the underwater topography data and the simulation precision requirements of the river and reservoir network, a one-dimensional hydrodynamic model and a two-dimensional hydrodynamic model are established, including: According to the characteristics of the river network, the river and reservoir network structure is established. According to the simulation precision requirements, a one-dimensional hydrodynamic model is established for the case where the longitudinal distance is greater than the horizontal and vertical directions, and a two-dimensional hydrodynamic model in the horizontal and vertical directions is established for lakes; The calculation formula of the one-dimensional hydrodynamic model is as follows: Where A is the cross-sectional area, t is the time, Q is the flow rate, x is the flow rate, q is the interval unit flow rate, g is the acceleration of gravity, B is the water surface width, Z is the water level, n is the roughness, and R is the wetted perimeter; The calculation formula of the two-dimensional hydrodynamic model is as follows: Where h is the water depth, u and v are the average velocity components of the water depth in the x and y directions respectively, and τ xx ,τ yx ,τ xy ,τ yy is the viscous stress, τ zxb ,τ zyb is the shear stress of the riverbed, τ zxs ,τ zys are the free surface stress, ρ is the density, and f is the Coriolis coefficient.

6. The device according to claim 5, characterized in that The basic geographic information of the basin includes digital elevation model data and digital river network data of the basin; the underlying surface data includes land use type data and soil type data; the underwater terrain data includes large-section data and underwater terrain point cloud data; the hydrological and meteorological data includes air pressure, temperature, precipitation, humidity, wind speed, sunshine data of meteorological stations in and around the basin and runoff observation data of hydrological stations; the water conservancy project data includes engineering parameter data of reservoirs, water intakes, drainage outlets, pumping stations, culverts, and flood storage areas in the basin.

7. The device according to claim 6, characterized in that The coefficient determination module is further configured to: Based on the digital elevation model of the watershed, the steepest slope method is used in combination with the digital river network geographic information data of the watershed to divide the sub-watersheds and define the river network and encode it to establish a watershed network map; Based on the watershed network diagram, a dynamic hydrological response unit is established taking into account the historical changes in underlying land use types and the spatial distribution of soil types. The dynamic hydrological response unit can describe the dynamic characteristics of the underlying surface under the changes in different types of land use; Based on the time-varying gain theory, precipitation, evaporation, runoff, soil water, and groundwater flow are calculated in each dynamic hydrological response unit. The calculation formula is as follows: P i +AW i =AW i+1 +RS i +ET i +RI i +RG i (1) Where, P i is the rainfall in the i-th period, AW i is the soil moisture content in the i-th period, ET i is the evapotranspiration of the i-th period, RS i is the surface runoff in the i-th period, RI i is the soil flow in the i-th period, RG i is the groundwater runoff in the i-th period, i is the period number, AW is the soil moisture content, P is the rainfall, f(AW) is the function expression, AW j+1 is the soil moisture value of the j+1th iteration, AW j is the soil moisture value of the jth iteration, f(AW j ) is the jth iteration function value, f′(AW j ) is the derivative value of the jth iteration; Where g1, g2 are time-varying gain runoff factors, WM is the maximum soil water content, K r ,K rg are the runoff coefficients of soil flow and groundwater runoff, α is the slope, L is the slope length, WFC is the field water holding capacity, AW u,i is the upper soil moisture content in the i-th period, AW d,i is the soil moisture content in the lower layer during the i-th period, AW d,i+1 is the soil moisture content in the lower layer during the i+1th period; Where E is evapotranspiration, T,E i ,E s are vegetation transpiration, canopy interception evaporation, and soil evaporation, respectively; min is the minimum function; λ is the latent heat of evaporation; Δ is the gradient of the saturated water vapor pressure-temperature curve; ρ is the air density; C p is the constant pressure specific heat of air, R nc ,R ns is the net radiation of the canopy and the surface, G is the soil heat flux, γ is the dry-bulb constant, f w is the wetted area ratio, r ac is the canopy surface boundary layer impedance, r as is the soil surface boundary layer impedance, r s is the soil impedance, r c is the canopy impedance, LW is the leaf water content, and D0 is the vapor pressure difference. 8 . A non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, executes the method according to claim 1 .

Images