A method for rapid prediction of urban flood based on multi-process generalization

Abstract

The present application relates to the technical field of flood forecasting, and more particularly to a city flood rapid forecasting method based on multi-process generalization. The method comprises the following steps: dividing sub-catchment by obtaining basic geographic information such as river system, terrain and rainwater pipe network of the target area, analyzing city underlying surface characteristics to calculate runoff, and calculating surface and pipe network confluence process based on improved equiflux line method and generalized trunk pipe model; through the interaction of trunk pipe and series-parallel virtual reservoir model, the surface detention, overland flow and return pipe are realized; finally, the one-dimensional river confluence and the cellular automata two-dimensional surface inundation model are coupled to realize the rapid simulation and visualization forecasting of city flood process. The present application efficiently couples runoff, surface confluence, pipe network flow limitation, surface and pipe network interaction, one-dimensional river confluence and two-dimensional inundation process through multi-process generalization, greatly reduces the calculation complexity and improves the timeliness of city flood forecasting.

Description

Technical Field

[0001] This invention relates to the field of flood forecasting technology, and in particular to a rapid urban flood forecasting method based on multi-process generalization. Background Technology

[0002] With increasing urbanization, flooding has become a prominent weakness affecting urban safety. Compared to natural watersheds, flood prevention in high-density cities is more challenging, mainly due to difficulties in rainfall forecasting, rapid flood runoff, and short defense windows. Furthermore, the high economic exposure and dense network of critical infrastructure in high-density cities make them more susceptible to secondary disasters and exacerbate losses. Therefore, urban flood forecasting, as a crucial non-engineering measure in emergency management, is an important means to enhance resilience and improve decision-making capabilities.

[0003] Although there are already a large number of research results and rich practical applications, various flood forecasting technologies have their own advantages and disadvantages, making it difficult to balance computational speed and simulation accuracy: traditional hydrological methods are not adaptable to urban areas, refined hydrological and hydrodynamic simulation methods are inefficient in meeting the needs of rapid response to urban floods and are difficult to obtain basic data, and data-driven forecasting methods suffer from a shortage of monitoring data. Summary of the Invention

[0004] Therefore, it is necessary to provide a rapid urban flood forecasting method based on multi-process generalization to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a rapid urban flood forecasting method based on multi-process generalization is proposed, the method comprising the following steps: Step S1: Obtain basic geographic information of the target area; divide the target area into sub-catchments based on the basic geographic information, which includes river system data, topographic data, and urban stormwater drainage network data; Step S2: Analyze the urban underlying surface type of the target area and calculate the total runoff of the sub-catchment areas; Step S3: Calculate the surface and pipe network confluence process of the sub-catchment area by improving the isochrone and one-dimensional trunk pipe model, and simulate the surface storage, overflow and surface backflow pipe process by using the one-dimensional trunk pipe model and virtual reservoir; Step S4: Calculate the river confluence process based on the one-dimensional river model; simulate the two-dimensional surface inundation process of the target area based on the virtual reservoir storage capacity and the two-dimensional surface inundation model, and visualize the simulation results to obtain urban flood forecast results.

[0006] The present invention has the following beneficial effects: By unifying and organically coupling multiple key hydrological and hydraulic processes in the formation mechanism of urban flooding, such as runoff generation, confluence, pipeline transportation and discharge, surface backflow, storage, and surface inundation, this method achieves rapid and efficient forecasting of urban flooding processes while ensuring the rationality of the physical mechanisms. On one hand, this method introduces isocurrent time zone division and virtual reservoir generalization mechanisms at the sub-catchment scale, transforming complex spatial confluence and surface water accumulation processes into computable and controllable parameterized processes. This not only significantly reduces the model's dependence on high-precision continuous hydrodynamic calculations but also effectively improves computational efficiency and stability in large-scale, complex urban areas. On the other hand, by comparing and analyzing the outflow scale and outflow timing characteristics of the first upstream sub-catchment with the drainage capacity parameters of the second downstream sub-catchment, and introducing clear numerical criteria (including peak flow ratio ranges and duration ratio ranges), it can quickly identify effective sub-catchments where surface backflow may occur in the early stages of rainfall, avoiding redundant calculations for areas without backflow risk, thereby further improving the overall forecasting speed and specificity. Furthermore, this method combines a one-dimensional trunk line model with a two-dimensional surface inundation model based on the concept of cellular automata. By fully considering the coupling effects of stormwater pipe network flow restriction, surface storage and backflow, it achieves an intuitive and visual representation of the urban flooding process. In addition, it improves the readability and application value of the forecast results in terms of spatial details through image resolution enhancement. Attached Figure Description

[0007] Figure 1 A flowchart illustrating the steps of a rapid urban flood forecasting method based on multi-process generalization; Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3. Figure 3 This is a flood forecast visualization diagram of a rapid urban flood forecasting method based on multi-process generalization proposed in this application; Figure 4 This is a schematic diagram of a one-dimensional trunk line model for a rapid urban flood forecasting method based on multi-process generalization proposed in this application. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0008] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0009] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0010] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0011] To achieve the above objectives, please refer to Figures 1 to 4 A rapid urban flood forecasting method based on multi-process generalization, the method comprising the following steps: Step S1: Obtain basic geographic information of the target area; divide the target area into sub-catchments based on the basic geographic information, which includes river system data, topographic data, and urban stormwater drainage network data; In one embodiment, basic geographic information of the target area is first acquired as the foundational data for subsequent analysis and modeling. This basic geographic information includes river system data, topographic data, and urban stormwater drainage network data, which may originate from geographic information systems, urban planning databases, or actual surveying and mapping results.

[0012] Among them, river system data is used to characterize the distribution pattern, flow direction relationship and confluence structure of natural water bodies in the target area, including the location of the river centerline, the connection relationship of river segments and the upstream and downstream topological relationship; topographic data is used to reflect the elevation distribution characteristics and surface undulation of the target area, which can reflect the natural slope aspect and convergence trend of surface water flow; urban stormwater pipe network data is used to describe the spatial layout and connectivity of artificial drainage systems, including the pipe direction, the location of inspection wells and the distribution of drainage outlets.

[0013] After obtaining the aforementioned basic geographic information, a preliminary hydrological spatial analysis of the target area is conducted. Specifically, by combining the elevation distribution characteristics in the topographic data, the main convergence direction of surface water flow and the location of watershed ridges are identified; at the same time, by combining river system data, the inflow paths and control ranges of natural water bodies are determined; and by combining urban stormwater drainage network data, the impact of artificial drainage systems on surface runoff paths is analyzed.

[0014] Based on the above analysis, the target area is spatially divided into several relatively independent sub-catchments using topographic watersheds, river control sections, and key nodes of the stormwater drainage network as boundaries. Each sub-catchment has a relatively consistent flow direction and drainage outlet, and its surface runoff can be collected to the corresponding downstream node through natural rivers or stormwater drainage networks.

[0015] In some embodiments, to improve the rationality of sub-catchment division, the connectivity and integrity of the preliminary division results can be checked to ensure that there is no overlap or unclear confluence path among the sub-catchments. For areas with ambiguous boundaries or unclear drainage paths, the boundaries can be locally corrected based on the surrounding topography and pipeline network orientation.

[0016] Finally, the sub-catchment division results of the target area are output, and corresponding spatial range information and basic attribute identifiers are generated for each sub-catchment.

[0017] Step S2: Analyze the urban underlying surface type of the target area and calculate the total runoff of the sub-catchment areas; In one embodiment, after the target area is divided into sub-catchments, the urban underlying surface types within each sub-catchment are analyzed to clarify the impact characteristics of different surface types on the rainfall runoff process. The urban underlying surface types include impermeable surfaces and permeable surfaces, wherein impermeable surfaces may include roads, squares, building roofs, etc., and permeable surfaces may include green spaces, bare soil, water bodies, and other surface types with permeability.

[0018] In the specific implementation process, based on existing land use data, remote sensing imagery data, or urban planning data, the land surface within the sub-catchment area is classified and identified, and the identification results are mapped to the corresponding sub-catchment area. Through spatial overlay analysis, the distribution of different underlying surface types in each sub-catchment area is statistically analyzed to obtain the proportion information of each type of land surface within the sub-catchment area.

[0019] For different underlying surface types, the runoff generation characteristics under rainfall conditions were analyzed. Impermeable surfaces, due to their weak permeability, are prone to surface runoff during rainfall; permeable surfaces, on the other hand, can reduce surface runoff through infiltration and retention. In the analysis, factors such as surface roughness, surface slope, and surface cover were comprehensively considered to differentiate the runoff generation contribution of various underlying surfaces.

[0020] In some embodiments, historical rainfall data of the target area can be combined to summarize and analyze the response characteristics of different underlying surface types under different rainfall intensities and durations, thereby improving the rationality of runoff estimation results. This approach enables the underlying surface analysis results to more accurately reflect the actual situation of urban rainfall runoff.

[0021] After completing the analysis of underlying surface types within the sub-catchment area, the runoff of each type of underlying surface is summarized and calculated to obtain the total runoff of the corresponding sub-catchment area. The total runoff is the sum of the surface runoff generated by each type of surface within the sub-catchment area during rainfall, and is used to characterize the runoff generation capacity of the sub-catchment area under the current rainfall conditions.

[0022] In some embodiments, to avoid excessive influence of local extreme surface types on the total runoff calculation results, the spatial continuity of underlying surface types within the sub-catchment area can be checked. When a certain surface type exhibits a highly fragmented distribution, its runoff contribution can be smoothed by combining adjacent surface types, thereby ensuring the stability of the total runoff calculation results.

[0023] Finally, the analysis results of the underlying surface type and the total runoff calculation results for each sub-catchment area are output, and these results are used as important input data for subsequent sub-catchment area runoff analysis and rapid urban flood forecasting.

[0024] Step S3: Calculate the surface and pipe network confluence process of the sub-catchment area by improving the isochrone and one-dimensional trunk pipe model, and simulate the surface storage, overflow and surface backflow pipe process by using the one-dimensional trunk pipe model and virtual reservoir; In one embodiment, the surface runoff process is calculated using an improved isohyet method based on topographic and rainfall data of the sub-catchment area. Specifically, the sub-catchment area is divided into multiple isohyet zones, each representing a regional unit with similar runoff times. For each isohyet zone, parameters such as topographic slope, flow direction, soil permeability, and underlying surface type are extracted to calculate the surface runoff generation and flow rate curves into the main drainage channel for that area. By superimposing the flow rate curves from each isohyet zone, the total surface runoff process of the entire sub-catchment area during rainfall is obtained.

[0025] Secondly, based on urban stormwater drainage network data, a one-dimensional trunk pipe model is constructed to simulate the network confluence process. In practice, information such as pipe diameter, slope, node locations, and stormwater well layout is input into the model to calculate the flow rate and water level changes of each pipe over different time periods. The water accumulation in overcapacity pipe sections is determined based on network capacity and flow restriction conditions. When the network capacity is insufficient, the overflow rate is calculated and fed back to the surface to reflect the surface return flow process.

[0026] Subsequently, the surface runoff and pipe network runoff processes are coupled with a virtual reservoir model to simulate the retention, overflow, and backflow processes in sub-catchments. In practice, each isocurrent time zone is generalized as a virtual reservoir unit, and the storage capacity and discharge conditions of the virtual reservoir are set according to the terrain elevation and drainage characteristics of the sub-area. Surface runoff is input into the virtual reservoir as the retention volume. Water exceeding the pipe network's processing capacity can flow into adjacent areas through overflow paths or return to the main pipe, achieving a realistic simulation of the surface retention and backflow pipeline processes.

[0027] During the simulation, the water level changes, pipeline backflow, and overflow of each virtual reservoir can be recorded in real time, forming comprehensive water volume evolution data for the sub-catchment areas. Through the above process, a complete confluence process, including surface runoff, pipeline flow, storage, and backflow, can be obtained, providing input data for subsequent urban flood forecasting.

[0028] Step S4: Calculate the river confluence process based on the one-dimensional river model; simulate the two-dimensional surface inundation process of the target area based on the virtual reservoir storage capacity and the two-dimensional surface inundation model, and visualize the simulation results to obtain urban flood forecast results.

[0029] In one embodiment, based on the obtained surface and pipeline confluence process of the sub-catchment area, the river confluence process is simulated using a one-dimensional trunk pipe model. The one-dimensional trunk pipe model considers factors such as river cross-section, riverbed slope, river roughness, and inflow rate to calculate the river water level and flow rate at each time step, so as to accurately reflect the hydrodynamic evolution process within the river.

[0030] For the surface inundation simulation of the target area, the surface is divided into two-dimensional grid cells, and two-dimensional hydrodynamic simulation is performed using the above-mentioned river confluence results and the retention, overflow, and backflow volumes of sub-catchments. During the simulation, topographic elevation, underlying surface type, river overflow points, and the backflow effect of urban drainage networks are considered to calculate the water depth and water accumulation range of each grid cell at different time steps.

[0031] In one embodiment, reference may be made to Figure 4 The one-dimensional main channel model uses the river centerline as its framework, dividing the river into multiple computational units. Within each unit, corresponding river cross-sectional attributes and upstream / downstream connections are associated, forming a complete river confluence calculation path. Darker colors in the main channel indicate higher flow rates. By using the inflow information determined in step S3 for the sub-catchments as boundary input conditions for the river model, the runoff from each sub-catchment can converge downstream according to the actual river system structure, thus obtaining the dynamic confluence process within the river channel.

[0032] Subsequently, while calculating river confluence, a two-dimensional surface inundation simulation model based on a surface spatial grid is introduced. This two-dimensional surface inundation model divides the surface into regular or irregular surface calculation units based on elevation data, road and building distribution information of the target area, and incorporates the surface backflow pipeline process and retention process information confirmed in step S3 into each surface calculation unit. When river water levels or pipe network drainage capacity are limited, the amount of water overflowing from the river or pipe network is mapped to the corresponding surface calculation unit, thereby triggering the simulation of surface water diffusion and spread.

[0033] In a preferred embodiment, reference may be made to Figure 3 Based on a two-dimensional surface grid of the study area, a surface inundation evolution model is constructed, and key linear constraint elements and point drainage elements are introduced into the surface grid. In the figure, the blue linear elements are used to represent roads, embankments or elevation change zones, which guide or block the surface runoff path and the range of water accumulation. The square point elements are used to represent storm drains, manholes or local low-lying units, and the water depth distribution at different times is represented by concentric color levels around them. During the calculation process, the two-dimensional surface inundation model and the one-dimensional trunk pipe model adopt a time-synchronous joint calculation method. Within the same time step, based on the flow capacity in the trunk pipe and the changes in node water level, the return flow, overflow, and local water accumulation depth of the corresponding surface unit are synchronously corrected. At the same time, the water accumulation depth distribution formed in the surface unit (as shown in the figure, inundation depth zones of different radii) inversely affects the inflow conditions of storm drains and inspection wells, thereby forming a dynamic feedback relationship between road constraints, drainage nodes, and surface water accumulation range. This allows the spatial differentiation characteristics of local high water accumulation areas, shallow water accumulation areas, and non-inundation areas to be accurately depicted over time, and more realistically reflects the formation and expansion process of urban flooding at the micro-topographic scale.

[0034] After completing the entire simulation calculation, the obtained results of river water level changes and two-dimensional surface water distribution are subjected to unified spatial mapping and visualization processing. Specifically, the surface water depth, water accumulation range, and duration information at different times are superimposed onto the base map of the target area to generate a flood evolution image or dynamic layer with time series characteristics. This is used to intuitively display the location, development process, and receding trend of urban flooding, thereby obtaining the final urban flood forecast results.

[0035] Preferably, the sub-catchment areas of the target region in step S1 based on basic geographic information include: Based on river system data and topographic data, the spatial orientation and distribution of rivers in the region are analyzed to obtain information on the river network structure of the target area. A digital elevation model of the target area is constructed using terrain data, and the set of terrain features corresponding to each location within the target area is calculated using the digital elevation model. Based on river network structure information and topographic feature set, the main channels of natural surface runoff and their watershed boundaries are identified, and the natural runoff zones within the target area are determined. The catchment area of ​​the target area is confirmed by using urban stormwater pipe network data, and the catchment area information is spatially matched with natural runoff zones to identify the sub-catchment areas of the target area.

[0036] In one embodiment, taking a built-up area of ​​a city as the target area, the river system data, topographic data and urban stormwater pipe network data corresponding to the area are first obtained. The river system data includes the river centerline, river level and connectivity information. The topographic data is raster or point cloud elevation data with a unified elevation benchmark. The urban stormwater pipe network data includes information such as stormwater pipeline direction, pipe diameter, manhole location and drainage outlet location.

[0037] By identifying the direction of the river centerline, its bifurcation relationships, and its upstream and downstream connections, the spatial distribution pattern of the river within the target area is extracted, and the main river channel and tributary channels are distinguished, thus forming complete river network structure information.

[0038] Subsequently, a digital elevation model (DEM) of the target area is constructed based on the acquired terrain data. By performing unified projection, outlier removal, and smoothing on the terrain data, a continuous and stable surface elevation representation is generated. Based on this, the DEM is analyzed to extract the set of terrain features corresponding to each location within the target area. This set of terrain features includes at least slope aspect characteristics, relative elevation relationships, and terrain undulation characteristics, used to reflect the movement trend of surface water under natural conditions.

[0039] Next, based on the river network structure information and topographic feature set, the main channels of natural surface runoff within the target area are identified. Specifically, by analyzing the relationship between topographic elevation changes and river channel spatial location, the paths through which surface runoff preferentially converges and is transported to the river channel under rainfall conditions are determined, and the watershed boundaries between different confluence paths are identified. This divides the target area into several natural confluence zones, ensuring that surface runoff within each natural confluence zone naturally flows towards the same river channel or river segment.

[0040] After completing the natural runoff zoning, urban stormwater network data is used to refine and detail the zoning. By analyzing the spatial orientation of stormwater pipelines and their correspondence with surface roads and plots, the actual catchment area served by each stormwater outlet and manhole is determined, generating corresponding catchment area information. Subsequently, this catchment area information is spatially matched with the aforementioned natural runoff zoning. When the catchment area spans multiple natural runoff zoning zones, the natural runoff boundaries are adjusted according to the drainage direction of the stormwater network, making the runoff boundaries more consistent with the actual operational characteristics of the urban artificial drainage system.

[0041] Ultimately, by combining the runoff characteristics controlled by the natural terrain and the drainage characteristics dominated by the urban stormwater pipe network, the target area is subdivided into multiple sub-catchments, each of which has a clear natural runoff direction and a defined urban drainage outlet.

[0042] As an example of the present invention, reference is made to Figure 2 As shown, step S3 in this example includes: Step S31: Calculate the total runoff time of the sub-catchment area based on the topographic data and urban stormwater pipe network data, and determine the corresponding outflow process based on the total runoff time; Step S32: Construct a one-dimensional trunk pipe model based on urban stormwater pipe network data, and simulate pipe network flow restriction based on the one-dimensional trunk pipe model; Step S33: Divide the sub-catchment area into multiple virtual reservoirs connected in series and parallel by dividing the equal flow time zone during the outflow process, and enter the virtual reservoir as the virtual reservoir storage volume by taking the excess water volume in the one-dimensional trunk pipe model that exceeds the flow limit of the main trunk pipe. Based on the hydraulic connection between the virtual reservoirs, the characteristics of the virtual reservoirs, and the flow limit of the pipe network, simulate the water exchange between the virtual reservoirs and the pipeline, and the water exchange between the virtual reservoirs to obtain the pipeline processes of stagnation, overflow and surface return.

[0043] In one embodiment, the total runoff time of the sub-catchment area is calculated based on topographic data (including elevation, slope, flow direction, etc.) and urban stormwater drainage network data (including pipe diameter, pipe length, slope, and node location, etc.). Specifically, the sub-catchment area is divided into several small grid units. Based on the slope and underlying surface permeability characteristics of each unit, the time required for surface runoff generated by rainfall to flow into the main drainage channel is calculated. The runoff times of each unit are then spatially accumulated and weighted to obtain the total runoff time of the sub-catchment area. Subsequently, an outflow process curve of the sub-catchment area is generated based on the total runoff time to describe the amount of water discharged into the drainage network or adjacent areas at each time point under rainfall.

[0044] Then, a one-dimensional trunk pipe model is constructed based on urban stormwater pipe network data. In practice, information such as pipe network nodes, pipe diameters, pipe slopes, and stormwater well layouts are input into the model, and the flow and water level changes of the pipe network at different time periods are simulated. During the simulation, considering pipe network capacity limitations, flow restriction calculations are performed on the main and branch pipes. When the pipe flow exceeds the design capacity, the excess water volume is recorded and used as a potential surface overflow or backflow input.

[0045] Next, by dividing the outflow process into isocurrent time zones, areas with similar hydrological characteristics within the sub-catchment are grouped into multiple sub-regions. Each sub-region is generalized as a virtual reservoir unit to simulate the surface retention effect. In practice, the excess water volume exceeding the pipeline flow limit in the one-dimensional main pipe model is input into the corresponding virtual reservoir as the storage volume. The virtual reservoir calculates the retained water volume and discharge volume at each time step based on the terrain elevation, storage capacity, and discharge conditions.

[0046] Hydraulic connections are established between virtual reservoirs to simulate the series and parallel water exchange between them. Simultaneously, the discharge flow of the virtual reservoirs is coupled with the pipeline water flow to obtain the pipeline return flow and surface runoff. Through this process, complete water evolution data encompassing storage, runoff, and surface runoff pipeline processes are ultimately generated, providing input information for urban flood forecasting.

[0047] Preferably, step S31 includes: The main surface runoff paths of rainwater into the rainwater pipe network are identified by topographic data, and the surface runoff time of rainwater entering the pipe network is calculated based on the length, slope and surface conditions of the surface runoff paths. Based on urban stormwater pipe network data, the transmission path of stormwater in sub-catchment areas is identified from the inlet to the main pipe via branch pipes. The branch pipe confluence time is calculated based on the branch pipe length, pipe diameter and hydraulic conditions. The surface runoff time and the branch pipe runoff time are combined to obtain the total runoff time corresponding to each location in the sub-catchment area. The total runoff time includes the surface runoff time of rainwater entering the pipe from the surface and the branch pipe runoff time of rainwater being transmitted to the main pipe through the branch pipe. Based on the total confluence time, isochrones of surface and pipe network confluence are drawn within the sub-catchment area, and the sub-catchment area is divided into several isochrone zones according to the isochrones; The area corresponding to each equal flow time zone is statistically analyzed to construct the convergence time-area sequence of the equal flow time zones. Each equal flow time zone corresponds to a set of convergence time and area parameters. By calculating the confluence time-area sequence of the isocurrent time zone through convolution, the flow process at the outflow outlet of the sub-catchment is obtained, thereby determining the outflow process of the sub-catchment. The convolution calculation formula is shown below: For the export section in time Traffic, For time Net rainfall intensity, For the first The area of ​​each isocurrent region For the first Convergence time for each region.

[0048] In one embodiment, taking a sub-catchment area of ​​a city where basic geographic information databases have been established as the object, the main runoff paths of rainwater on the surface are first identified based on the topographic data of the area. By analyzing the surface elevation undulations, the natural trend of rainwater flowing from high to low elevations under rainfall conditions is determined. Combined with road alignment, green space distribution, and hardened surface conditions, the main runoff channels from the surface into the stormwater drainage network are selected. Based on this, considering the actual length of each runoff path, ground slope, and the influence of different underlying surfaces on the rainwater flow velocity, the time required for rainwater to flow along the surface and ultimately enter the stormwater drainage network is determined, thereby obtaining the spatial distribution characteristics of surface runoff time.

[0049] In this embodiment, the transmission process of rainwater within the sub-catchment area after entering the network from each inlet is analyzed by further integrating urban stormwater network data. By identifying the connection relationships between each stormwater inlet and its corresponding branch pipe, the transmission path of rainwater from the branch pipe to the main pipe is determined. For different branch pipe sections, considering their length, pipe diameter, and water delivery conditions under operating conditions, the transport characteristics of rainwater within the branch pipe are evaluated, and the time taken for rainwater to travel from entering the network to reaching the main pipe is determined accordingly. Through the above analysis, the corresponding branch pipe confluence time can be obtained for different locations within the sub-catchment area.

[0050] Based on this, the surface runoff time and the branch pipe runoff time are integrated to obtain the total runoff time distribution of rainwater at various locations within the sub-catchment area from its fall to its entry into the main pipe. According to this total runoff time distribution, time zone boundaries reflecting differences in runoff response are constructed within the sub-catchment area, and isochrones with the same runoff time characteristics are drawn spatially. Through the spatial distribution of the isochrones, the sub-catchment area is divided into multiple isochrone time zones with different runoff response characteristics, ensuring that rainwater within each isochrone time zone exhibits relatively consistent runoff characteristics over time.

[0051] In this embodiment, the coverage area of ​​each isocurrent time zone is statistically analyzed, and the runoff time and area information corresponding to each isocurrent time zone are compiled to form a time-area correspondence reflecting the overall runoff structure of the sub-catchment area. By comprehensively superimposing the runoff contributions of each isocurrent time zone in different time periods, the gradual collection of rainwater during rainfall can be simulated, thereby determining the flow change process of the sub-catchment outflow during the entire rainfall duration. This outflow process can realistically reflect the order in which rainwater from different areas arrives at the outflow and its impact on the overall outflow.

[0052] Preferably, step S33 includes: By dividing the isocurrent time zones during the outflow process, the isocurrent time zones with similar hydrological characteristics in the sub-catchment area are merged and divided to obtain multiple sub-regions. Each sub-area is generalized as a virtual reservoir, and the surface retention effect of the sub-area is simulated by the virtual reservoir to obtain the retention process. The excess water volume in the one-dimensional trunk pipe model that exceeds the flow limit of the main trunk pipe is entered into the virtual reservoir as the virtual reservoir storage volume. Based on the topographic data and storage process corresponding to the isocurrent time zone, the water level-storage capacity relationship of each virtual reservoir is constructed. Based on the topographic elevation and rainwater well distribution of the main pipeline in the sub-area, the water level-discharge relationship of each virtual reservoir is constructed; The calculation formula for the backflow main pipe of the virtual reservoir in the sub-area is as follows: In the formula, This refers to the surface water volume of the return main pipeline. The maximum flow capacity of the main pipe. for Main pipe flow rate at all times for Real-time virtual reservoir discharge capacity; Surface backflow and overflow are calculated based on the virtual reservoir's water storage capacity, water level-storage capacity relationship, water level-discharge relationship, and series-parallel relationship between virtual reservoirs to obtain the overflow and surface backflow pipeline process of the sub-catchment area.

[0053] In one embodiment, the isocurrent time zones divided during the outflow process are processed. Specifically, isocurrent time zones with similar hydrological characteristics (e.g., areas with similar slope, land cover type, and underlying surface permeability coefficient) within the sub-catchment area are merged and divided into multiple sub-regions. Each sub-region is treated as a separate unit in the simulation to maintain consistency in hydrological characteristics.

[0054] Subsequently, each sub-area is generalized into a virtual reservoir. For each virtual reservoir, the storage process is obtained by simulating the surface water retention effect. In practice, the excess water volume exceeding the main pipeline flow limit in the one-dimensional trunk pipeline model is input into the corresponding virtual reservoir as the water storage volume, so that the virtual reservoir can reflect the surface water accumulation caused by pipeline capacity limitations.

[0055] Next, based on the topographic elevation and water storage process corresponding to the isochronous time zone of each sub-region, a virtual reservoir water level-storage capacity relationship is constructed, so that the water level change of each reservoir corresponds to the water storage capacity. Based on the topographic elevation of the main pipeline and the distribution of rainwater wells in the sub-region, a virtual reservoir water level-discharge relationship is further constructed to simulate the process of the virtual reservoir discharging water into the main pipeline.

[0056] The following method is used to calculate the surface return water volume in the main pipeline: First, the main pipeline flow rate and maximum flow capacity at time t are obtained, and the corresponding discharge capacity of the virtual reservoir at the same time point is also obtained. If the main pipeline flow rate exceeds the pipeline capacity, the surface return water volume is taken as the minimum value between the remaining flow rate of the main pipeline and the discharge capacity of the virtual reservoir, thus obtaining the sub-area return flow rate.

[0057] Finally, by combining the virtual reservoir's water storage capacity, water level-storage capacity relationship, water level-discharge relationship, and the series and parallel hydraulic connections between virtual reservoirs, the surface backflow and overflow of the sub-catchment area are calculated, obtaining the complete overflow and surface backflow pipeline process of the sub-catchment area, providing input data for further urban flood forecasting.

[0058] Preferably, each sub-area is generalized as a virtual reservoir, and the surface water retention effect of the sub-area is simulated through the virtual reservoirs to obtain the water retention process, which includes: Each sub-region is generalized as a virtual reservoir. Virtual reservoirs within the same sub-region are connected in series according to the hydrological connection between upstream and downstream areas, while virtual reservoirs in adjacent sub-regions are connected in parallel. The surface water storage effect of a sub-region is simulated using a virtual reservoir to obtain the storage process.

[0059] In one embodiment, each sub-area within the sub-catchment area is generalized as a virtual reservoir to simulate the surface water retention effect of the sub-area. The capacity of the virtual reservoir is related to the area, topographic elevation, and underlying surface characteristics of the sub-area to ensure that the simulated water level changes match the actual surface water storage situation.

[0060] In some embodiments, for virtual reservoirs within the same sub-area, they are connected in series according to the hydrological connections and drainage direction within the sub-area, following the hydraulic sequence of upstream and downstream, to ensure that water volume is transferred sequentially along the main flow direction and to realize the simulation of continuous storage process.

[0061] In some embodiments, virtual reservoirs in adjacent sub-areas are connected in parallel, so that the water stored in adjacent sub-areas can simultaneously enter the corresponding pipe network or downstream sub-area, simulating the spatial distribution and confluence effect of water.

[0062] In some embodiments, based on the water level-capacity relationship and water level-discharge relationship of the virtual reservoir, the water level of each virtual reservoir is calculated over time to obtain the time series of water storage volume in the sub-area, thereby completely describing the surface storage process of the sub-catchment area and providing a data basis for further flooding and backflow simulation.

[0063] Preferably, step S4 includes: A one-dimensional river model is constructed based on the acquired river cross-section information. The outflow from the main pipe and the flow from the surface to the river are used as the input boundary of the one-dimensional river model. The river confluence process is simulated through the one-dimensional river model to obtain the river confluence process. Based on the sub-regional topography, a two-dimensional surface water inundation model is automatically constructed using cells; A two-dimensional surface water inundation model and virtual reservoir water retention capacity are used to simulate the two-dimensional surface inundation process of the target area; The two-dimensional surface inundation process and river confluence process are visualized, and the image resolution of the visualized two-dimensional surface inundation process and river confluence process is enhanced to obtain urban flood forecast results.

[0064] In one embodiment, a one-dimensional river channel model is constructed in a computer environment based on acquired river cross-sectional information (such as river width, cross-sectional shape, and riverbed elevation). In a specific implementation, the outflow from the main pipe of the sub-catchment area and the flow from surface runoff into the river channel are used as the input boundary conditions for the one-dimensional river channel model. The river confluence process is simulated using the one-dimensional river channel model to obtain the flow changes and water level evolution of the target area river channel at different time steps. This simulation process can be implemented using explicit or implicit difference methods to ensure the numerical stability and accuracy of the simulation.

[0065] Meanwhile, considering the terrain of the sub-regions, the target area is divided into multiple two-dimensional cells through spatial discretization, and a two-dimensional surface water inundation model is constructed based on the cellular automata method. In specific implementation, the state of each cell is represented by the water depth at a preset time step, and by defining the cell neighborhood relationship and water exchange rules, the migration and distribution of water between adjacent cells are realized, thereby simulating the diffusion process of surface flood.

[0066] After obtaining the water storage volume of the virtual reservoir in the sub-catchment area, it is used as the input of the two-dimensional surface water inundation model. At the same time, combined with the surface storage, overflow and surface return pipeline processes, a two-dimensional flooding simulation is performed on the target area to obtain the changes in surface water level and water depth.

[0067] Finally, the data obtained from step S3, including surface retention, overflow, and surface backflow processes, as well as the two-dimensional surface inundation and river confluence processes, are visualized. In implementation, 3D rendering technology can be used to present surface water depth and river levels as color gradients or contour surfaces. Simultaneously, image resolution enhancement is applied to the visualization results to ensure that details such as flood extent and river level changes are clearly visible. Through these steps, complete urban flood forecasting results can be obtained, which can be used for flood early warning and drainage scheduling decisions.

[0068] Preferably, based on the sub-regional topography, the automatic construction of a two-dimensional surface water inundation model using cells includes: Based on the topography of the sub-region, the target area is spatially discretized and divided into multiple spatial discrete units, where each spatial discrete unit serves as a cell in the two-dimensional surface water inundation model. The water depth in each cell at a preset time point is taken as the cell state, and the cell states at each time point are used to form a state set. For any cell, other cells that have a water exchange relationship with it are defined as the neighborhood of that cell, and the neighborhood range is defined by at least one of the Moore-type neighborhood and Magorean-type neighborhood to obtain the cell neighborhood relationship. Based on the cell neighborhood relationship, the water exchange rule between cells is defined as the migration rule, and water is allocated between adjacent cells through the migration rule to construct a two-dimensional surface water inundation model.

[0069] In one embodiment, the target area is spatially discretized based on the terrain data of the sub-area. Specifically, terrain elevation data within the sub-area is acquired, and the target area is divided into grids according to a uniform spatial resolution, dividing the entire area into multiple regular spatial discrete units, such as square or rectangular grids; each spatial discrete unit corresponds to a small piece of land surface in the target area and serves as a cell in a two-dimensional surface water inundation model, used to independently describe the water accumulation state at that location.

[0070] Secondly, the water depth within each cell at a preset time point is used as the cell state. Specifically, at the initial moment of the model, the water depth of each cell is set to zero or assigned a value based on the initial rainfall conditions. During model operation, the water depth within each cell is updated at fixed time steps, and the water depth state of all cells at different time points is recorded, thus forming a set of cell states reflecting the evolution of surface water accumulation. For example, during continuous rainfall, the cell state at each time point can correspond to a snapshot of surface water distribution.

[0071] Subsequently, for any given cell, other cells that exchange water with it are defined as its neighborhood. Specifically, based on the cell's positional relationship in the spatial grid, a Moore-type neighborhood method is used to determine its eight surrounding neighboring cells, or a Magorecian-type neighborhood method is used to group neighboring cells to adapt to the water diffusion requirements under different terrain conditions. Through the above methods, the neighborhood range within the two-dimensional plane where each cell can exchange water is clarified, thereby establishing a complete cell neighborhood relationship.

[0072] Based on this, water exchange rules between cells are defined as migration rules based on the cell neighborhood relationships. Specifically, the direction and priority of water transfer between adjacent cells are set by combining the terrain elevation difference between adjacent cells, the current water depth, and spatial connectivity. In each time step, the water volume between adjacent cells is allocated and updated according to the migration rules, so that water can diffuse from higher cells to lower cells, or form local water accumulation in relatively enclosed areas. By continuously executing the above water migration process, a two-dimensional surface water inundation model is gradually constructed and run, thereby simulating the surface water diffusion and inundation process of sub-regions.

[0073] Preferably, the rules for water exchange between cells are defined as migration rules, including: Between a cell and its neighboring cells, hydraulic parameters for water migration calculation are determined based on the spatial scale characteristics of the cell. The cell side length is used as the hydraulic radius parameter, and the hydraulic gradient in the direction of water transfer is determined based on the difference in water surface elevation between adjacent cells and the distance between cell centers. Based on hydraulic parameters and a preset time step, the scale of water migration from a cell to its neighboring cells within a single time step is determined, and the migration volume is characterized in the form of water depth change. Based on the cellular neighborhood relationship, water balance constraints and neighborhood stability constraints are introduced to the migration water volume to obtain the migration rules.

[0074] In one embodiment, based on a spatially discretized urban surface cellular grid, for any target cell, the spatial relationship characteristics between it and its neighboring cells are first analyzed. According to a unified spatial discretization standard, the side length of each cell is used as a hydraulic characteristic parameter reflecting the spatial scale of that cell, characterizing the diffusion capacity of water within and at the cell boundary. Simultaneously, by combining the current water surface elevation of the target cell and its neighboring cells, the relative water surface elevations are compared, and the actual distance between cell centers is considered to determine the dominant direction of potential water migration, thus clarifying the directional basis for water transfer.

[0075] In this embodiment, when the water surface elevation of the target cell is higher than that of a neighboring cell, it is determined that the neighborhood direction meets the conditions for water transfer. At this point, based on the aforementioned hydraulic characteristic parameters and a pre-set time step, the potential scale of water migration within a single time step is estimated. The migrated water volume is not directly expressed as volume, but rather converted into the change in water depth between the target cell and neighboring cells. This describes the impact of water transfer on the water accumulation state of each cell, facilitating unified updates within the cell state set.

[0076] Furthermore, to avoid abnormal concentration or unreasonable amplification of local water volume during the water migration process, a water balance constraint mechanism is introduced in this embodiment. Specifically, the water volume migrating from the target cell to multiple neighboring cells is uniformly constrained to ensure that the total outflow after migration does not exceed the effective water volume that the target cell can release within the current time step. At the same time, the water receiving capacity of neighboring cells is constrained to prevent non-physical abrupt water accumulation due to excessive inflow in a short period of time.

[0077] Through the combined effects of determining the direction and scale of water migration, as well as constraints on water balance and stability, a complete cellular water exchange and migration rule is constructed. During the operation of the two-dimensional surface water inundation model, the migration rule repeatedly acts on all cells at each time step, enabling surface water to diffuse and flow back in an orderly manner in space and gradually form stagnant water in low-lying areas, thus more realistically reflecting the evolution of surface water accumulation under urban rainfall conditions.

[0078] Preferably, before dividing the sub-catchment area into multiple virtual reservoirs connected in series and parallel by dividing the isocurrent time zones during the outflow process, the following steps are also included: Based on the connection relationship of urban stormwater pipe network data and the topographic flow direction, the hydraulic relationship between the first sub-catchment area and the second sub-catchment area is determined. The first sub-catchment area is located upstream of the second sub-catchment area or shares a drainage channel with the second sub-catchment area. Based on the total runoff and runoff time characteristics of the first sub-catchment area, the outflow scale and outflow timing characteristics of the downstream drainage channel of the first sub-catchment area are determined. Based on the urban stormwater pipe network structure corresponding to the second sub-catchment area, determine the drainage capacity parameters of the second sub-catchment area; Based on the outflow scale and timing characteristics of the first sub-catchment area and the drainage capacity parameters of the second sub-catchment area, determine whether the second sub-catchment area has surface backflow triggering conditions. If so, mark it as a valid sub-catchment area; otherwise, mark it as an invalid sub-catchment area and remove it.

[0079] In one embodiment, based on urban stormwater network data of the target area, the pipe connections between various sub-catchments are analyzed. By analyzing the upstream and downstream connection sequence of stormwater pipes, pipe diameter changes, and node confluence relationships, and combining this with the natural confluence direction reflected in the topographic data, it is determined whether there is a clear hydraulic transfer path between different sub-catchments. For sub-catchments located upstream of the drainage path, or those sharing the same drainage trunk line or key nodes with downstream sub-catchments, a hydraulic correlation is determined to exist between them, thus forming a clear upstream-downstream sub-catchment combination.

[0080] Based on this, the runoff generation and confluence characteristics of the first sub-catchment located upstream are analyzed. Specifically, considering the area characteristics, topographic slope, and internal drainage path structure of this sub-catchment, its total runoff generation capacity and confluence response speed under rainfall conditions are comprehensively evaluated, and the overall outflow scale to the downstream drainage channel is determined accordingly. Furthermore, the temporal distribution characteristics of this outflow are analyzed to identify whether the outflow exhibits strong concentration, long duration, or prominent peak temporal characteristics.

[0081] Subsequently, for the second sub-catchment area, which is hydraulically connected to the first sub-catchment area, the corresponding urban stormwater pipe network structure was analyzed in detail. By examining the layout of main and branch pipes, drainage path length, key node locations, and distribution of local low-lying areas within this sub-catchment area, the drainage capacity level of this sub-catchment area under the combined effects of rainfall and upstream water inflow was comprehensively determined. This drainage capacity parameter is used to reflect the overall capacity boundary of the second sub-catchment area's effective drainage volume per unit time.

[0082] Based on the above analysis, the outflow scale and temporal characteristics of the first sub-catchment are compared with the drainage capacity of the second sub-catchment. When the outflow intensity or duration of the upstream sub-catchment exceeds the drainage capacity of the downstream sub-catchment within a specific time period, the second sub-catchment is determined to have the triggering conditions for surface backflow and is marked as an effective sub-catchment, included in the subsequent isocurrent time zone division and virtual reservoir construction process. Conversely, when the downstream sub-catchment has sufficient drainage capacity and can promptly absorb the upstream water, it is determined not to have the triggering conditions for surface backflow and is marked as an invalid sub-catchment, removed from the overall modeling process to reduce model complexity and improve simulation efficiency.

[0083] Of particular importance is that, based on the urban stormwater drainage network structure corresponding to the second sub-catchment, the drainage capacity parameters of the second sub-catchment are determined as follows: Based on the spatial boundary of the second sub-catchment area, determine the scope of the urban stormwater pipe network that is connected to the second sub-catchment area for drainage. Within the urban stormwater pipe network, identify the main drainage channels that undertake the drainage function of the second sub-catchment area. The main drainage channels include at least branch pipe sections and their corresponding trunk pipe sections that are consistent with the outflow direction of the second sub-catchment area. Extract the pipe diameter, pipe segment length, slope, and connectivity of the main drainage channels as structural parameters; Based on structural parameters, the single-segment drainage capacity of each main drainage pipe section is determined, and combined with the series or parallel relationship between pipe sections, the effective drainage capacity under the corresponding pipe network structure of the second sub-catchment area is obtained. Effective drainage capacity is used as the drainage capacity parameter for the second sub-catchment area.

[0084] In one embodiment, based on the spatial boundary of the second sub-catchment area, the rainwater discharge relationship within and at its boundary is analyzed, with a focus on identifying urban stormwater network units that have direct drainage connections with this sub-catchment area. Through a comprehensive analysis of the location of stormwater wells, pipeline routes, and network topology, the actual scope of the urban stormwater network responsible for the drainage of the second sub-catchment area is determined, thereby avoiding the inclusion of pipe segments unrelated to this sub-catchment area in the calculation.

[0085] Within the defined urban stormwater drainage network, the main drainage channels that play a dominant role in draining the second sub-catchment area are further identified. These main drainage channels are preferably combinations of pipe sections that align with the surface runoff direction of the second sub-catchment area and preferentially collect water from the sub-catchment area during rainfall. Specifically, these include key branch pipe sections located downstream of the sub-catchment area, and corresponding trunk pipe sections connected to these branch sections and undertaking the runoff collection function. This method accurately reflects the main transport path of stormwater from the sub-catchment area after entering the network.

[0086] Subsequently, structural parameters were extracted from each pipe segment in the main drainage channel. These extracted parameters included the pipe diameter, segment length, laying slope, and connection method between segments. These structural parameters comprehensively characterize the drainage potential of each pipe segment under actual operating conditions and its adaptability to changes in inflow water.

[0087] Based on this, the drainage capacity of each major drainage pipe section under normal operating conditions is evaluated segment by segment according to the structural parameters of each section. The overall drainage capacity is then comprehensively adjusted by considering the series or parallel connections between pipe sections. This approach avoids overestimating or underestimating the capacity of a single pipe section, thus more accurately reflecting the overall drainage level of the second sub-catchment area under the constraints of the existing pipe network structure.

[0088] Finally, the comprehensive drainage capacity obtained under this pipeline structure will be used as the drainage capacity parameter of the second sub-catchment area, and will be used for subsequent comparative analysis with the outflow scale and time sequence characteristics of the upstream first sub-catchment area to determine whether the triggering conditions for surface backflow are met.

[0089] Of particular importance is that, based on the spatial boundary of the second sub-catchment area, the scope of the urban stormwater drainage network that connects with the second sub-catchment area includes: The spatial boundary information of the second sub-catchment area is obtained, and the spatial boundary is used to define the surface water catchment range of the second sub-catchment area. Within the spatial boundary, identify storm drains, storm wells or other surface drainage facilities located in the second sub-catchment area as drainage interfaces between the second sub-catchment area and the urban stormwater pipe network. Starting from the drainage interface, along the connection direction of the urban stormwater pipe network, trace the stormwater pipe segments that are directly or indirectly connected to the drainage interface to form a set of candidate pipe network units. The candidate pipeline unit set is screened, and pipeline units that are not connected to the second sub-catchment area in the drainage direction are removed, while pipeline units that can carry the outflow of the second sub-catchment area are retained. The selected set of pipe network units is defined as the urban stormwater pipe network range that is connected to the second sub-catchment area for drainage.

[0090] In one embodiment, based on the division results of the second sub-catchment area, its corresponding spatial boundary information is obtained. The spatial boundary is based on the topographic watershed line, road boundary or artificial drainage boundary, and is used to clarify the surface water catchment control range of the second sub-catchment area.

[0091] Based on this, within the spatial boundary, various surface drainage facilities installed in the second sub-catchment area are identified one by one. These surface drainage facilities include storm drains, storm wells, linear drainage ditches, or other drainage structures connected to the urban stormwater pipe network. These drainage facilities, as key nodes for surface runoff entering the urban stormwater pipe network, are uniformly identified as the drainage interface between the second sub-catchment area and the urban stormwater pipe network, used to characterize the actual inlet location of surface water entering the pipe network system.

[0092] Subsequently, using each drainage interface as a starting node, and based on the topological connections of the urban stormwater pipe network, stormwater pipe segments directly connected to the drainage interfaces or indirectly connected via multiple pipe sections are traced and identified along the drainage direction of the network. Through this tracing process, all pipe segments that may carry water from the second sub-catchment area are included in the analysis scope, forming a candidate pipe network unit set. This candidate pipe network unit set can comprehensively reflect the possible transport paths of stormwater from the second sub-catchment area after entering the pipe network.

[0093] After forming a set of candidate network units, the candidate network units are further screened based on the actual drainage direction of the network, the slope of the pipe sections, and the downstream inflow relationship. Network units that are spatially connected but cannot carry the outflow of the second sub-catchment in the drainage direction, such as pipe sections with opposite drainage directions or those that only undertake the drainage task of other sub-catchments, are eliminated; while network units that can continuously receive and transport the outflow of the second sub-catchment in the drainage path are retained.

[0094] Ultimately, the selected network units were uniformly defined as the urban stormwater network area that connects with the second sub-catchment area for drainage.

[0095] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0096] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A rapid urban flood forecasting method based on multi-process generalization, characterized in that, Includes the following steps: Step S1: Obtain basic geographic information of the target area; The target area is divided into sub-catchments based on basic geographic information, which includes river system data, topographic data, and urban stormwater drainage network data. Step S2: Analyze the urban underlying surface type of the target area and calculate the total runoff of the sub-catchment areas; Step S3: Calculate the surface and pipe network confluence process of the sub-catchment area by improving the isochrone and one-dimensional trunk pipe model, and simulate the surface storage, overflow and surface backflow pipe process by using the one-dimensional trunk pipe model and virtual reservoir; Step S4: Calculate the river confluence process based on a one-dimensional river model; The simulation of the two-dimensional surface inundation process in the target area is based on the virtual reservoir water storage capacity and the two-dimensional surface inundation model. The simulation results are then visualized to obtain urban flood forecast results.

2. The rapid urban flood forecasting method based on multi-process generalization according to claim 1, characterized in that, Step S1, which involves dividing the target area into sub-catchments based on basic geographic information, includes: Based on river system data and topographic data, the spatial orientation and distribution of rivers in the region are analyzed to obtain information on the river network structure of the target area. A digital elevation model of the target area is constructed using terrain data, and the set of terrain features corresponding to each location within the target area is calculated using the digital elevation model. Based on river network structure information and topographic feature set, the main channels of natural surface runoff and their watershed boundaries are identified, and the natural runoff zones within the target area are determined. The catchment area of ​​the target area is confirmed by using urban stormwater pipe network data, and the catchment area information is spatially matched with natural runoff zones to identify the sub-catchment areas of the target area.

3. The rapid urban flood forecasting method based on multi-process generalization according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Calculate the total runoff time of the sub-catchment area based on the topographic data and urban stormwater pipe network data, and determine the corresponding outflow process based on the total runoff time; Step S32: Construct a one-dimensional trunk pipe model based on urban stormwater pipe network data, and simulate pipe network flow restriction based on the one-dimensional trunk pipe model; Step S33: Divide the sub-catchment area into multiple virtual reservoirs connected in series and parallel by dividing the equal flow time zone during the outflow process, and enter the virtual reservoir as the virtual reservoir storage volume by taking the excess water volume in the one-dimensional trunk pipe model that exceeds the flow limit of the main trunk pipe. Based on the hydraulic connection between the virtual reservoirs, the characteristics of the virtual reservoirs, and the flow limit of the pipe network, simulate the water exchange between the virtual reservoirs and the pipeline, and the water exchange between the virtual reservoirs to obtain the pipeline processes of stagnation, overflow and surface return.

4. The rapid urban flood forecasting method based on multi-process generalization according to claim 3, characterized in that, Step S31 includes: The main surface runoff paths of rainwater into the rainwater pipe network are identified by topographic data, and the surface runoff time of rainwater entering the pipe network is calculated based on the length, slope and surface conditions of the surface runoff paths. Based on urban stormwater pipe network data, the transmission path of stormwater in sub-catchment areas is identified from the inlet to the main pipe via branch pipes. The branch pipe confluence time is calculated based on the branch pipe length, pipe diameter and hydraulic conditions. The surface runoff time and the branch pipe runoff time are combined to obtain the total runoff time corresponding to each location in the sub-catchment area. The total runoff time includes the surface runoff time of rainwater entering the pipe from the surface and the branch pipe runoff time of rainwater being transmitted to the main pipe through the branch pipe. Based on the total confluence time, isochrones of surface and pipe network confluence are drawn within the sub-catchment area, and the sub-catchment area is divided into several isochrone zones according to the isochrones; The area corresponding to each equal flow time zone is statistically analyzed to construct the convergence time-area sequence of the equal flow time zones. Each equal flow time zone corresponds to a set of convergence time and area parameters. By calculating the confluence time-area sequence of the isocurrent time zone through convolution, the flow process at the outflow outlet of the sub-catchment is obtained, thereby determining the outflow process of the sub-catchment. The convolution calculation formula is shown below: For the export section in time Traffic, For time Net rainfall intensity, For the first The area of ​​each isocurrent region For the first Convergence time for each region.

5. The rapid urban flood forecasting method based on multi-process generalization according to claim 3, characterized in that, Step S33 includes: By dividing the isocurrent time zones during the outflow process, the isocurrent time zones with similar hydrological characteristics in the sub-catchment area are merged and divided to obtain multiple sub-regions. Each sub-area is generalized as a virtual reservoir, and the surface retention effect of the sub-area is simulated by the virtual reservoir to obtain the retention process. The excess water volume in the one-dimensional trunk pipe model that exceeds the flow limit of the main trunk pipe is entered into the virtual reservoir as the virtual reservoir storage volume. Based on the topographic data and storage process corresponding to the isocurrent time zone, the water level-storage capacity relationship of each virtual reservoir is constructed. Based on the topographic elevation and rainwater well distribution of the main pipeline in the sub-area, the water level-discharge relationship of each virtual reservoir is constructed; The calculation formula for the backflow main pipe of the virtual reservoir in the sub-area is as follows: In the formula, This refers to the surface water volume of the return main pipeline. The maximum flow capacity of the main pipe. for Main pipe flow rate at all times for Real-time virtual reservoir discharge capacity; Surface backflow and overflow are calculated based on the virtual reservoir's water storage capacity, water level-storage capacity relationship, water level-discharge relationship, and series-parallel relationship between virtual reservoirs to obtain the overflow and surface backflow pipeline process of the sub-catchment area.

6. The rapid urban flood forecasting method based on multi-process generalization according to claim 5, characterized in that, Each sub-region is generalized as a virtual reservoir, and the surface water retention effect of the sub-region is simulated using virtual reservoirs to obtain the water retention process, which includes: Each sub-region is generalized as a virtual reservoir. Virtual reservoirs within the same sub-region are connected in series according to the hydrological connection between upstream and downstream areas, while virtual reservoirs in adjacent sub-regions are connected in parallel. The surface water storage effect of a sub-region is simulated using a virtual reservoir to obtain the storage process.

7. The rapid urban flood forecasting method based on multi-process generalization according to claim 1, characterized in that, Step S4 includes: A one-dimensional river model is constructed based on the acquired river cross-section information. The outflow from the main pipe and the flow from the surface to the river are used as the input boundary of the one-dimensional river model. The river confluence process is simulated through the one-dimensional river model to obtain the river confluence process. Based on the sub-regional topography, a two-dimensional surface water inundation model is automatically constructed using cells; A two-dimensional surface water inundation model and virtual reservoir water retention capacity are used to simulate the two-dimensional surface inundation process of the target area; The two-dimensional surface inundation process and river confluence process are visualized, and the image resolution of the visualized two-dimensional surface inundation process and river confluence process is enhanced to obtain urban flood forecast results.

8. The rapid urban flood forecasting method based on multi-process generalization according to claim 7, characterized in that, Based on the sub-regional topography, a two-dimensional surface water inundation model is automatically constructed using cells, including: Based on the topography of the sub-region, the target area is spatially discretized and divided into multiple spatial discrete units, where each spatial discrete unit serves as a cell in the two-dimensional surface water inundation model. The water depth in each cell at a preset time point is taken as the cell state, and the cell states at each time point are used to form a state set. For any cell, other cells that have a water exchange relationship with it are defined as the neighborhood of that cell, and the neighborhood range is defined by at least one of the Moore-type neighborhood and Magorean-type neighborhood to obtain the cell neighborhood relationship. Based on the cell neighborhood relationship, the water exchange rule between cells is defined as the migration rule, and water is allocated between adjacent cells through the migration rule to construct a two-dimensional surface water inundation model.

9. The rapid urban flood forecasting method based on multi-process generalization according to claim 8, characterized in that, The rules for water exchange between cells are defined as migration rules, including: Between a cell and its neighboring cells, hydraulic parameters for water migration calculation are determined based on the spatial scale characteristics of the cell. The cell side length is used as the hydraulic radius parameter, and the hydraulic gradient in the direction of water transfer is determined based on the difference in water surface elevation between adjacent cells and the distance between cell centers. Based on hydraulic parameters and a preset time step, the scale of water migration from a cell to its neighboring cells within a single time step is determined, and the migration volume is characterized in the form of water depth change. Based on the cellular neighborhood relationship, water balance constraints and neighborhood stability constraints are introduced to the migration water volume to obtain the migration rules.

10. The rapid urban flood forecasting method based on multi-process generalization according to claim 3, characterized in that, Before dividing the sub-catchment area into multiple virtual reservoirs connected in series and parallel by isocurrent time zones during the outflow process, the following steps are also included: Based on the connection relationship of urban stormwater pipe network data and the topographic flow direction, the hydraulic relationship between the first sub-catchment area and the second sub-catchment area is determined. The first sub-catchment area is located upstream of the second sub-catchment area or shares a drainage channel with the second sub-catchment area. Based on the total runoff and runoff time characteristics of the first sub-catchment area, the outflow scale and outflow timing characteristics of the downstream drainage channel of the first sub-catchment area are determined. Based on the urban stormwater pipe network structure corresponding to the second sub-catchment area, determine the drainage capacity parameters of the second sub-catchment area; Based on the outflow scale and timing characteristics of the first sub-catchment area and the drainage capacity parameters of the second sub-catchment area, determine whether the second sub-catchment area has surface backflow triggering conditions. If so, mark it as a valid sub-catchment area; otherwise, mark it as an invalid sub-catchment area and remove it.

Images