Method and system for constructing overflow facilities in sponge cities based on high-precision rain and flood simulation

Through high-precision rainfall simulation and ecological walking space analysis, the planning and design of sponge urban overflow facilities is optimized, which solves the problems of low simulation accuracy and insufficient optimization in the existing technology, and achieves more efficient flood control and ecological collaborative optimization.

CN118607251BActive Publication Date: 2024-11-05CHINA INST OF WATER RESOURCES & HYDROPOWER RES
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410823110.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-25
Publication Date
2024-11-05
Estimated Expiration
2044-06-25

AI Technical Summary

Technical Problem

The planning and design of existing sponge urban overflow facilities has problems such as low simulation accuracy and insufficient facility optimization, and it is difficult to accurately characterize the spatial and temporal changes of urban rainfall and flooding processes and to comprehensively consider facility site selection, connectivity and ecological benefits on the urban scale. Multi-target factor.

Method used

Using high-precision rainfall simulation method, a variety of overflow area planning schemes were generated by extracting hydrological meteorological data, constructing three-dimensional topographic models, coupled surface-pipe drainage models, combining ecological network and pedestrian channel analysis, and performing fine simulations and risk assessment, optimize overflow facility parameters to achieve systematic and collaborative layout.

Benefits of technology

The rationality of overflow facilities layout and flood prevention and disaster reduction effects are improved, taking into account urban ecological restoration and residents' leisure and recreation needs, and the multi-objective coordinated optimization of sponge city construction has been achieved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118607251B_ABST
    Figure CN118607251B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for constructing overflow facilities in a sponge city based on high-precision rain and flood simulation, comprising the following steps: determining the scope of the urban area, collecting basic data and preprocessing; extracting hydrological and meteorological data, calling a pre-constructed rain and flood simulation model to analyze the water accumulation situation, and preliminarily extracting and screening overflow areas according to the water accumulation situation, and constructing an overflow area set; forming an overflow area set with connectivity information, and generating at least two overflow area planning schemes; for each overflow area planning scheme in the overflow area set with connectivity information, using a rain and flood simulation model for refined simulation, and giving overflow facility parameters under typical rain and flood conditions, forming an overflow facility parameter set. The quality and efficiency of overflow facility construction are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to technologies related to sponge city construction, and in particular to a method and system for constructing overflow facilities in a sponge city based on high-precision rain and flood simulation. Background Art

[0002] Sponge city is an important concept and strategy to deal with urban waterlogging and water environment deterioration. It aims to improve the city's water ecological resilience and sustainable development capabilities by building a multi-level urban water management system of "infiltration, retention, storage, purification, use, and discharge". In the construction of sponge cities, overflow facilities are key facilities for controlling urban runoff and reducing peak flow. They are of great significance for preventing urban waterlogging disasters and maintaining a healthy water environment. However, the current planning and design of sponge city overflow facilities still face many technical challenges.

[0003] To this end, some research has been conducted on the overflow facilities of sponge cities at home and abroad. The first existing technology analyzes the overflow characteristics of low-impact development facilities under different return period rainfall based on the SWMM model, providing a basis for the planning and design of overflow facilities. The second existing technology uses the Infoworks ICM model to evaluate the control effect of different overflow outlet layout schemes on urban waterlogging and optimize the spatial configuration of overflow facilities. The third existing technology couples the one-dimensional pipe network model and the two-dimensional surface flow model to simulate the urban overflow process, and determines the elevation and size of the overflow weir accordingly.

[0004] Existing technologies have promoted the planning and design of sponge city overflow facilities to a certain extent, but there are still problems such as low simulation accuracy and insufficient facility optimization. On the one hand, existing studies mostly use simplified hydrological and hydrodynamic models, which are difficult to accurately describe the spatiotemporal dynamic changes of urban rain and flood processes, resulting in a lack of refined hydrological basis for the planning and design of overflow facilities. On the other hand, existing studies on the optimization of overflow facilities are mostly limited to a single facility scale, lacking comprehensive consideration of multi-objective factors such as facility site selection, connectivity, and ecological benefits at the urban scale, making it difficult to achieve systematization and coordination of the layout and construction of overflow facilities.

[0005] Therefore, further research and development and innovation are needed in order to solve the problems existing in the existing technology. Summary of the invention

[0006] The purpose of the invention is to provide a method and system for constructing sponge city overflow facilities based on high-precision rain and flood simulation, in order to solve one of the above-mentioned technical problems.

[0007] Technical solution, according to one aspect of the present application, a method for constructing a sponge city overflow facility based on high-precision rain and flood simulation includes the following steps:

[0008] Step S1, determine the scope of the urban area, collect basic data and pre-process;

[0009] Step S2: based on the preprocessed basic data, extract the hydrological and meteorological data, call the pre-built rain and flood simulation model to analyze the waterlogging situation, and preliminarily extract and screen the overflow area according to the waterlogging situation to construct an overflow area set;

[0010] Step S3: based on the preprocessed basic data, extract the ecological pedestrian channel and construct the existing ecological and pedestrian orientation map; for each overflow area, extract the newly added ecological and pedestrian orientation sub-map, and superimpose it with the existing ecological and pedestrian orientation map to form an overflow area set with connectivity information, and generate at least two overflow area planning schemes;

[0011] Step S4: for each overflow area planning scheme in the overflow area set with connectivity information, a rainstorm simulation model is used to perform a refined simulation, and overflow facility parameters under typical rainstorm conditions are given to form an overflow facility parameter set.

[0012] According to another aspect of the present application, a sponge city overflow facility construction system based on high-precision rain and flood simulation is also provided, including:

[0013] at least one processor; and,

[0014] a memory communicatively connected to at least one of the processors; wherein,

[0015] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the method for constructing sponge city overflow facilities based on high-precision rain and flood simulation as described in any of the above technical solutions.

[0016] Beneficial effect: first, through the simulation model, the overflow area is quickly screened out, and then the planning scheme is generated by combining the ecological and pedestrian guide map, and then the scheme is screened. After the screening, a refined simulation is performed, and then the overflow facility parameters are given. By first optimizing the overflow area and then optimizing the parameters of the overflow device, better results are achieved and the speed is faster. The relevant technical effects will be explained below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of the present invention.

[0018] Figure 2 It is a flow chart of step S1 of the present invention.

[0019] Figure 3 It is a flow chart of step S2 of the present invention.

[0020] Figure 4 It is a flow chart of step S3 of the present invention.

[0021] Figure 5 It is a flow chart of step S4 of the present invention. DETAILED DESCRIPTION

[0022] According to one aspect of the present application, a method for constructing a sponge city overflow facility based on high-precision rain and flood simulation includes the following steps:

[0023] Step S1, determine the scope of the urban area, collect basic data and pre-process;

[0024] Step S2: based on the preprocessed basic data, extract the hydrological and meteorological data, call the pre-built rain and flood simulation model to analyze the waterlogging situation, and preliminarily extract and screen the overflow area according to the waterlogging situation to construct an overflow area set;

[0025] Step S3: based on the preprocessed basic data, extract the ecological pedestrian channel and construct the existing ecological and pedestrian orientation map; for each overflow area, extract the newly added ecological and pedestrian orientation sub-map, and superimpose it with the existing ecological and pedestrian orientation map to form an overflow area set with connectivity information, and generate at least two overflow area planning schemes;

[0026] Step S4: for each overflow area planning scheme in the overflow area set with connectivity information, a rainstorm simulation model is used to perform a refined simulation, and overflow facility parameters under typical rainstorm conditions are given to form an overflow facility parameter set.

[0027] By collecting high-precision multi-source data such as terrain, underlying surface, hydrology and meteorology, and conducting systematic preprocessing and quality control, reliable data support is provided for subsequent analysis. In particular, the method of constructing a three-dimensional terrain model using high-resolution DEM can more realistically depict the ups and downs of the urban surface and improve the accuracy of stormwater simulation. Advanced hydrological and hydrodynamic models and multi-scenario analysis techniques are used. Through the coupled surface-pipeline drainage model, the spatiotemporal variability of factors such as rainfall, underlying surface, and pipeline network is fully considered, and the entire process of urban stormwater formation, confluence, and discharge can be dynamically simulated. At the same time, by setting rainfall scenarios with different recurrence periods and combining waterlogging depth thresholds and terrain factors, urban flood-prone areas and key overflow nodes can be identified more comprehensively and objectively. The refinement level of simulation and identification has been improved, laying a solid hydrological foundation for the subsequent layout of overflow facilities.

[0028] Through the superposition analysis of urban spatial elements such as ecological networks and slow-moving systems, the ecological value and service functions of the overflow area in terms of biodiversity protection and residents' recreation were explored, and the organic integration of stormwater management facilities and urban green infrastructure was achieved. At the same time, quantitative indicators such as ecological circulation potential and walking gravitational potential were innovatively proposed, and graph theory, fuzzy mathematics and other methods were used to evaluate the importance of overflow nodes, breaking through the limitations of the previous single flood control goal. In the site selection of overflow facilities, multi-objective optimization, risk analysis, multi-criteria decision-making and other technologies were comprehensively used to scientifically balance the goals of peak reduction and waterlogging reduction, ecological restoration, and engineering costs, and obtain the Pareto optimal layout plan that takes into account both benefits and risks. These innovative practices reflect the systematic transformation of sponge city planning and design from "either flood control or ecology" to "both flood control and ecology".

[0029] In this application, since the hydrological and hydrodynamic model is used for stormwater simulation, the two concepts are not strictly distinguished. The constructed hydrological and hydrodynamic model is used to simulate stormwater, so it can also be called a stormwater simulation model.

[0030] In summary, in this embodiment, by making full use of advanced technologies such as high-precision rain and flood simulation, ecological and pedestrian space analysis, and multi-scenario risk assessment, the layout rationality of overflow facilities and the actual flood prevention and disaster reduction effects can be significantly improved, while taking into account urban ecological restoration and residents' leisure and recreation needs, and realizing multi-objective coordinated optimization of sponge city construction.

[0031] According to one aspect of the present application, step S1 specifically comprises:

[0032] Step S11, determine the scope of the urban area and collect basic data, the basic data includes urban DEM data, urban basic information, hydrological and meteorological data and ecological environment data; urban basic information includes land use status map and impervious surface distribution data;

[0033] Step S12: convert the format of the basic data, unify the coordinate system and spatial resolution, obtain the processed basic data and store it;

[0034] Step S13: construct a three-dimensional terrain model based on the processed basic data to provide terrain boundary conditions for stormwater simulation.

[0035] By collecting high-precision DEM, land use, impervious surface and other multi-source data, and performing pre-processing such as format conversion, coordinate unification and quality inspection, high-quality terrain and underlying surface input conditions are provided for building a refined stormwater simulation model. The three-dimensional terrain model can provide more refined and realistic terrain boundary conditions for stormwater simulation, improving the reliability and application value of the simulation results.

[0036] According to one aspect of the present application, step S2 further comprises:

[0037] Step S21, extracting the pre-processed hydrological and meteorological data to generate rainfall scenario data with different return periods;

[0038] Step S22: for each rainfall scenario data, call the pre-built rain and flood simulation model to preliminarily simulate the generation and convergence process of surface runoff, as well as the flow process of rainwater on the surface and in the pipe network, and analyze the depth and range of water accumulation in each area;

[0039] Step S23: extracting areas where the water depth exceeds the threshold as potential overflow areas, and screening and merging the potential overflow areas in combination with terrain and underlying surface factors to obtain a preliminary overflow area set.

[0040] By using pre-processed hydrological and meteorological data and calling pre-built rain and flood simulation models to conduct multi-scenario analysis, urban flood-prone areas and overflow areas can be accurately identified. By extracting waterlogging depth thresholds and screening terrain underlying surface factors, the overflow area set can be further optimized, providing key candidate areas for the subsequent overflow facility layout, reducing blindness and arbitrariness.

[0041] According to one aspect of the present application, step S3 further comprises:

[0042] Step S31, extracting the urban ecological network and the urban pedestrian network respectively based on the preprocessed basic data;

[0043] Step S32: read the urban ecological network and the urban pedestrian network and analyze the connectivity to obtain the existing ecological and pedestrian maps;

[0044] Step S33: for each overflow area, extract green space, water system, road and trail respectively, construct a new ecological sub-graph and a new walking sub-graph, analyze the connectivity, obtain the new ecological and walking graphs, and make topological association with the existing ecological and walking graphs, and extract the topological association information;

[0045] Step S34: Add the topological association information to the overflow area set to form an overflow area set containing ecological and pedestrian connectivity information.

[0046] Step S3, based on the identification of overflow areas, fully explores and utilizes the connectivity information of the ecological and pedestrian spaces within the city. By constructing the existing ecological and pedestrian guide map, the spatial distribution and connection status of the urban green infrastructure are identified; then new ecological and pedestrian sub-maps are extracted for the overflow area, and topological associations are superimposed with the existing guide map to form a complete overflow area ecological and pedestrian network. This process can fully utilize the ecological potential of the overflow area, incorporate it into the urban green infrastructure system, and achieve the coupling of multiple benefits such as stormwater management, ecological restoration, and leisure and recreation. At the same time, by generating a variety of overflow area planning schemes, more options are provided for subsequent scheme evaluation and optimization.

[0047] According to one aspect of the present application, step S4 is further:

[0048] Step S41: calling the rain and flood simulation model, performing a refined simulation for each overflow area planning scheme, obtaining the water accumulation depth, range and duration of the overflow area under each rainfall scenario; determining the key water accumulation nodes and confluence paths of the overflow area;

[0049] Step S42: Optimize the design parameters of the overflow facilities for the key water accumulation nodes of each overflow area under each rain and flood situation to obtain a parameter combination;

[0050] Step S43: using a risk analysis method to evaluate the peak shaving and flood reduction effect and failure risk of the overflow facilities under each parameter combination; generating an optimal overflow facility parameter set for each overflow area in each overflow area planning scheme, and attaching corresponding risk assessment information;

[0051] Step S44: construct an overflow area planning scheme evaluation model, evaluate the optimal overflow facility parameter set, and obtain a facility parameter set that meets preset conditions as the overflow setting parameter set.

[0052] Based on the overflow area planning scheme, a high-precision rain and flood simulation model is used to carry out detailed simulation, which can obtain important information such as key waterlogging nodes and confluence paths, and provide accurate hydrological and hydraulic basis for parameter optimization of overflow facilities. At the same time, the risk analysis method is used to evaluate the peak shaving and flood reduction effects and failure risks under different parameter combinations, and a planning scheme evaluation model is constructed to select the optimal parameter set, which can effectively balance the functionality and safety requirements of overflow facilities and improve the reliability and stability of their actual operation.

[0053] According to one aspect of the present application, in step S32, the process of reading the urban ecological network and the urban pedestrian network and analyzing the connectivity to obtain the existing ecological map is specifically as follows:

[0054] Step S321, obtaining pre-processed basic data, obtaining influencing parameters including terrain undulation, land use type, traffic roads and habitat quality, and constructing the resistance surface of the urban blue-green space network based on the influencing parameters;

[0055] Step S322: Taking each blue-green space patch as a source, calculate the minimum cumulative resistance distance from it to other blue-green space patches to generate a cost distance matrix; calculate the connectivity index of the blue-green space network based on the cost distance matrix, obtain the connectivity matrix and analyze the vulnerability of the network connectivity; the connectivity index includes the overall connectivity and the key patch connectivity;

[0056] Step S323: Based on the connectivity matrix, the Kruskal algorithm is used to iteratively select the edge with the minimum cost, generate the minimum ecological corridor and perform conflict detection to obtain the minimum ecological corridor network; the nodes of the minimum ecological corridor network are used as seed points, and the weighted Voronoi diagram is generated in combination with the road network data;

[0057] Step S324: Calculate the ecological connectivity and accessibility between each Voronoi cell and adjacent Voronoi cells in the weighted Voronoi diagram.

[0058] By comprehensively considering the influencing parameters such as terrain undulation, land use, traffic roads, and habitat quality to construct the resistance surface, and using the minimum cumulative resistance model and the improved Kruskal algorithm to extract the key corridors, the connectivity characteristics of the urban ecological network can be more comprehensively and objectively characterized, providing a basis for the improvement of the ecological function and corridor optimization of the overflow area. At the same time, the vulnerability of network connectivity can be analyzed to predict the ecological risks of the overflow area under extreme events, providing a reference for the formulation of corresponding early warning and emergency strategies.

[0059] According to one aspect of the present application, in step S33, the process of obtaining a newly added ecological and walking graph and topologically associating it with the existing ecological and walking graph is specifically as follows:

[0060] Step S331: abstract the overflow area set into an undirected graph, and calculate the spatial distance matrix between the overflow areas based on the undirected graph and use it as the edge weight;

[0061] Step S332: Use the Louvain algorithm to perform community detection on the graph G, identify closely connected overflow area communities, and form an overflow area community set; for each overflow area community, construct an overflow area subgraph within it;

[0062] Step S333: for each overflow area community, superimpose its corresponding overflow area sub-graph and the influence parameter to form a blue-green space network at the community scale;

[0063] Step S334: On the blue-green space network at the community scale, the minimum cumulative resistance model is used to calculate the connectivity matrix, and the improved Kruskal algorithm is used to extract the minimum ecological corridor to form an ecological orientation map at the community scale; and combined with the road network data and the residents' outdoor activity preference data, the weighted Voronoi partition is used to generate a pedestrian walking orientation map that takes into account ecological connectivity and accessibility, and a new ecological and walking map is obtained;

[0064] Step S335: splice and optimize the existing ecological and pedestrian map and the newly added ecological and pedestrian map to form an overall urban-scale ecological and pedestrian map.

[0065] By using the Louvain algorithm to identify closely connected overflow area communities and constructing a blue-green space network at the community scale, the ecological connectivity of the overflow area can be analyzed more specifically, reducing the boundary effect and scale dependence problems of large-scale analysis. On this basis, the minimum cumulative resistance model and the improved Kruskal algorithm are used to extract community-scale ecological corridors, and pedestrian walking guide maps are generated in combination with road networks and residents' activity preference data, achieving accurate matching and optimal configuration of the ecosystem service functions of the overflow area and the leisure and recreation needs of residents, reflecting the transformation of the overflow facility layout from "gray-based" to "gray-green combined".

[0066] According to one aspect of the present application, step S41 further includes evaluating the importance of the overflow area, specifically:

[0067] Step S411, extracting an ecological orientation map, and extracting an ecological attribute feature vector for each node in the ecological orientation map, wherein the node includes an overflow area; the ecological attribute feature vector includes a green area, a water area, and a biodiversity index;

[0068] Step S412, calculating the connectivity and betweenness centrality of the nodes, and using the ecological flow potential evaluation model, using the weighted geometric mean method to integrate the ecological attributes, connectivity and betweenness centrality of the nodes to obtain the ecological flow potential score of each node;

[0069] Step S413, extracting a pedestrian walking guide map, and extracting outdoor activity preference feature vectors including visit frequency, stay time, and activity type diversity from each node in the pedestrian walking guide map;

[0070] Step S414: Calculate the connectivity and load capacity of the node, adopt the walking gravitational potential evaluation model, and use Choquet integration to perform nonlinear fusion of the node's outdoor activity preference, connectivity, and load capacity to obtain a walking gravitational potential score for each node;

[0071] Step S415: construct a fuzzy decision matrix, where the rows represent overflow areas, the columns represent two decision criteria, namely, ecological circulation potential and pedestrian gravitational potential, and the matrix elements are interval fuzzy numbers; considering the relative importance of ecological benefits and pedestrian activity benefits, determine the weight vector of the decision criteria;

[0072] Step S416, calculate the distance of each overflow area relative to the positive ideal solution and the negative ideal solution, and use the Hamming distance to measure the difference between interval fuzzy numbers; calculate the comprehensive score of each overflow area, weighing its superiority relative to the positive ideal solution and its inferiority relative to the negative ideal solution; sort the overflow areas according to the comprehensive score to obtain the overflow area importance parameter.

[0073] Evaluating the importance of overflow area nodes from two dimensions, ecological circulation potential and walking gravity potential, can more comprehensively and systematically consider the ecological value and recreational value of overflow areas. Ecological circulation potential comprehensively evaluates the ecological attributes of nodes from aspects such as green area, water area, species diversity, connectivity, and betweenness centrality, while walking gravity potential evaluates the recreational attractiveness of nodes from aspects such as visit frequency, residence time, activity diversity, connectivity, and load capacity. The comprehensive evaluation of the two can reveal the key position and role of overflow areas in the urban ecological security pattern and public space system, and provide a decision-making basis for the priority layout and optimal management of subsequent overflow facilities. At the same time, the multi-criteria decision-making model is constructed using interval fuzzy numbers and Choquet integrals, which can effectively deal with the uncertainty problem in node importance evaluation and improve the reliability and interpretability of the evaluation results.

[0074] In another embodiment of the present application, according to one aspect of the present application, in step S23, the process of extracting the area where the water depth exceeds the threshold as the potential overflow area is specifically as follows:

[0075] Step S2301: Based on the rain and flood simulation results of step S22, extract the information such as water depth, flow rate, and flooded area at different time periods to construct a rain and flood dynamic evolution data set;

[0076] Step S2302: clustering the rain and flood dynamic evolution data using ST-DBSCAN, automatically identifying the waterlogged areas clustered in the spatiotemporal dimension, and extracting their dynamic boundaries by combining methods such as run-length encoding;

[0077] Step S2303: Perform morphological analysis on the dynamic waterlogging area boundary to extract dynamic change characteristics of the overflow area, such as maximum flooding range, waterlogging time, waterlogging area change rate, etc., as dynamic indicators for overflow area screening;

[0078] Step S2304: Combine dynamic indicators and static indicators (such as terrain and underlying surface) and use fuzzy comprehensive evaluation and other methods to dynamically evaluate and optimize the initial overflow area to obtain the final overflow area set.

[0079] By introducing spatiotemporal data mining technology, this optimization scheme can more accurately characterize the dynamic evolution of the overflow area and improve the pertinence and effectiveness of identification and screening.

[0080] According to one aspect of the present application, in step S23, the process of screening and merging potential overflow areas also includes optimizing the screening of overflow areas, specifically:

[0081] Step S231, constructing an overflow area planning scheme evaluation model, wherein the model optimization objectives include peak shaving and flood reduction effects, ecological connectivity, and engineering costs;

[0082] Step S232: using a genetic algorithm to solve the optimization model and obtain a Pareto optimal solution set for the overflow facility layout;

[0083] Step S233, evaluating the ecological connectivity improvement effect of overflow facility layout under different Pareto optimal solutions, and selecting the layout plan with the best ecological benefit;

[0084] Step S234: When in use, make a decision on the overflow facility layout plan, use methods such as hierarchical analysis to determine the weights of different factors, and rank and optimize the alternative layout plans.

[0085] On the basis of screening and merging potential overflow areas, the overflow area planning scheme is further optimized. By constructing a multi-objective evaluation model, trade-offs and coordination can be made between objectives such as peak load reduction and flood reduction, ecological connectivity, and engineering costs. Intelligent optimization technologies such as genetic algorithms are used to search for the Pareto optimal solution set, which can efficiently solve complex multi-objective combination optimization problems and obtain ideal overflow facility layout schemes. When evaluating and making decisions, the comprehensive use of hierarchical analysis and other methods can systematically consider the relative importance of different influencing factors and improve the scientificity and reliability of decision-making. The entire process implements the idea of ​​"problem-oriented and goal-driven", which fully reflects the leap of overflow area planning from "static site selection" to "dynamic optimization".

[0086] In a further embodiment, in step S3, the process of overflow area connectivity analysis and planning scheme generation may also be:

[0087] Based on the preprocessed basic data, the urban ecological network and urban pedestrian network are extracted respectively;

[0088] Read the urban ecological network and urban pedestrian network and analyze the connectivity to obtain the existing ecological and pedestrian maps;

[0089] The set of preliminary overflow areas identified in step S2 is recorded as O1. For each overflow area in O1, the green space, water system, roads and trails inside and in the neighborhood are extracted respectively, and a new ecological subgraph and a new pedestrian subgraph are constructed. The connectivity is analyzed to obtain a new ecological and pedestrian graph.

[0090] Topologically associate the newly added ecological and pedestrian maps with the existing ecological and pedestrian maps, extract the topological association information, and append it to the overflow area set O1 to form the overflow area set O2 containing ecological and pedestrian connectivity information;

[0091] Based on the overflow area set O2, at least two overflow area planning schemes A1 and A2 are generated, each scheme contains an overflow area subset;

[0092] A preliminary evaluation was conducted on the overflow area planning schemes A1 and A2, with evaluation indicators including ecological connectivity and walkability. The planning scheme A with better evaluation results was screened out, and the overflow area subset contained in it was recorded as O3.

[0093] In another embodiment of the present application, in step S4, the process of overflow facility parameter optimization and feedback adjustment may also be:

[0094] For the overflow area subset O3 selected in step S3, a rainstorm simulation model is used to perform a refined simulation to obtain information such as the water accumulation depth, range, and duration of each overflow area under different rainfall scenarios, and to determine key water accumulation nodes and confluence paths;

[0095] For the key waterlogging nodes in each overflow area, the design parameters of the overflow facilities are optimized to obtain a set of parameter combinations; the risk analysis method is used to evaluate the peak shaving and flood reduction effect and failure risk of the overflow facilities under each parameter combination;

[0096] Generate the optimal overflow facility parameter set for each overflow area, and add the corresponding risk assessment information to form the overflow facility parameter set P of the overflow area subset O3;

[0097] Feeding back the overflow facility parameter set P to step S3, the overflow area planning scheme A is updated and optimized; specifically, the overflow areas in the planning scheme A are screened and adjusted according to the risk assessment information in the parameter set P, and overflow areas with too high risks or too low benefits are eliminated to form an updated overflow area planning scheme A';

[0098] Repeat the above steps until the planning scheme A' meets the preset risk threshold and benefit requirements; finally, the optimized overflow area planning scheme A and the corresponding overflow facility parameter set P are obtained.

[0099] The overflow area planning scheme A and overflow facility parameter set P obtained in the above are used as input to construct an overflow facility layout optimization model. The objective function includes peak shaving and flood reduction benefits, ecological benefit improvement, and engineering cost.

[0100] A multi-objective optimization algorithm (such as NSGA-II) is used to solve the layout optimization model and obtain a set of Pareto optimal solutions, each of which corresponds to an overflow facility layout plan;

[0101] Evaluate the comprehensive benefits of the overflow facility layout scheme corresponding to each Pareto optimal solution, and select the layout scheme B* with the best comprehensive benefits;

[0102] Generate a visualization rendering and decision report of layout plan B* as the final result of the planning and design of sponge city overflow facilities.

[0103] at least one processor; and,

[0104] a memory communicatively connected to at least one of the processors; wherein,

[0105] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the method for constructing sponge city overflow facilities based on high-precision rain and flood simulation as described in any of the above technical solutions.

[0106] In summary, in the links of connectivity analysis, node evaluation, multi-objective decision-making, etc., through the flexible application and innovative combination of advanced methods such as Louvain community detection, minimum cumulative resistance model, interval fuzzy Choquet integral, genetic algorithm, etc., the coupling potential of urban water-ecology-activity space can be fully tapped, and the layout of overflow facilities can be transformed and upgraded from a single goal to multiple benefits, from rigid flood control to flexible management, and from engineering facilities to natural ecology. This is of great significance for improving the systematicity, synergy, resilience and sustainability of sponge city construction. By constructing ecological and pedestrian orientation maps, the overflow area is incorporated into the urban green infrastructure and public space network, and through methods such as topological association and graph analysis, the key nodes and important connections of the overflow area in the ecological security pattern and slow-moving system are quantitatively evaluated. This process breaks through the limitations of the single facility scale in traditional research, and examines the site selection and layout of overflow facilities from multiple perspectives such as the overall ecology, recreation, and transportation of the city, which is conducive to the coordinated optimization of facility functions and urban multiple needs. In the connectivity analysis and node importance assessment, the scheme also cleverly uses graph theory, ecology and multi-criteria evaluation methods such as Louvain community detection, minimum cumulative resistance model, Choquet integral, etc. to quantitatively judge and optimize the ecosystem service function and recreational competence of the overflow area, further enriching the target system and constraint conditions for the optimization of the overflow facility layout. Through high-temporal and spatial resolution rainfall simulation and pipe network hydrodynamic simulation, the dynamic change process of water accumulation in the overflow area under different rainfall scenarios can be accurately obtained, providing a more accurate and reliable hydrological and hydraulic basis for the subsequent parameter optimization and layout optimization of overflow facilities.

[0107] In some embodiments, basic urban information includes: high-precision digital elevation model (DEM) with a resolution of not less than 1 meter; high-resolution remote sensing imagery with a resolution of not less than 0.5 meters; three-dimensional urban model, including buildings, roads, green spaces and other elements; urban underground pipe network data, including rainwater pipe network, sewage pipe network, integrated pipe corridor, etc.; current land use map and planning map; impervious surface distribution data; hydrometeorological data: long-term rainfall data with a time resolution of not less than 10 minutes; evaporation data; hydrological soil type distribution data; underlying surface roughness data; ecological environment data: urban green space distribution data, including parks, greenways, etc.; water system distribution data, including rivers, lakes, wetlands, etc.; animal and plant distribution data, especially the habitat distribution of key protected species; acquisition of socio-economic data: population distribution data, including permanent population and floating population; economic development data, such as GDP, industrial structure, etc.; traffic flow data, including road flow, bus passenger flow, etc.

[0108] In some embodiments, an improved Huff model may also be used to simulate the depth and range of urban waterlogging. The specific process is as follows:

[0109] Based on historical rainfall and waterlogging data, the shape parameters and scale parameters of the Huff curve are estimated to characterize the rainfall time distribution pattern under different rainfall scenarios.

[0110] For each potential waterlogging point, the improved Huff formula is used to calculate the waterlogging depth under different rainfall scenarios, taking into account factors such as its catchment area, underlying surface infiltration loss, and pipe network drainage capacity.

[0111] Taking potential waterlogging points as seed points, cellular automata were used to simulate the spatial expansion process of waterlogging based on terrain undulation and hydraulic connectivity to obtain the waterlogging range.

[0112] By changing parameters such as rainfall intensity and duration, multiple rainfall scenarios are generated, and the changes in waterlogging depth and range under different scenarios are analyzed.

[0113] In another embodiment of the present application, the steps of simulating the rainfall process are specifically as follows:

[0114] Considering the spatial and temporal distribution and variability of rainfall, a rainfall process with high spatial and temporal resolution is generated.

[0115] The double-layer grid method is used to discretize the surface, the spatial heterogeneity of the underlying surface is considered, and the finite volume method is used to solve the two-dimensional shallow water equation to simulate the generation and confluence of surface runoff.

[0116] The one-dimensional Saint-Venant equation is used to describe the water flow in the pipe network, taking into account the connectivity and confluence direction of the pipes, and simulating the hydraulic transmission and discharge process of the pipe network.

[0117] The dynamic coupling of surface runoff and pipe network flow is achieved through exchange terms, and processes such as rainwater infiltration, overflow, and discharge are considered to improve the authenticity of the simulation.

[0118] The model is calibrated and verified using historical monitoring data to evaluate the simulation accuracy and stability of the model.

[0119] Simulate the rainstorm and flood processes under different return periods and rainfall types, and evaluate the waterlogging risk and drainage capacity of potential overflow areas.

[0120] In this example, the selection of key parameters of the hydrological and hydrodynamic model is crucial to the model performance. The main considerations include:

[0121] Topographic parameters, including elevation data accuracy, spatial resolution, river cross-section shape, etc. Soil parameters, including permeability, porosity, initial water content, etc. Subsurface parameters, including impervious area ratio, Manning roughness coefficient, etc. Rainfall parameters, including rainfall intensity, duration, spatial distribution, etc. Hydrological parameters, including evaporation, infiltration, groundwater recharge, etc. Hydraulic parameters, including flow rate, water depth, boundary conditions, etc.

[0122] Among them, parameter selection needs to consider: sensitivity analysis, identification of parameters that have a greater impact on model results. The physical meaning and applicable scope of the parameters. The difficulty and cost of obtaining the parameters. The temporal and spatial variability of the parameters.

[0123] In another embodiment of the present application, a three-level model structure is constructed, including:

[0124] Macro scale: watershed hydrological model; meso scale: urban hydrodynamic model; micro scale: detailed model of key nodes.

[0125] Determine the coupling relationship between models: from macro to meso, boundary conditions are transferred; from meso to micro, local conditions are extracted; from micro to meso, feedback correction is made.

[0126] Macro-scale model construction: collect basin DEM, land use, soil type data, organize meteorological data (rainfall, temperature, evaporation, etc.); collect hydrological station observation data

[0127] Based on DEM, sub-basin division is performed to extract river network structure.

[0128] Select a predetermined hydrological model (such as SWAT, VIC, etc.), set model parameters, and perform parameter calibration and verification.

[0129] Simulate rainfall-runoff processes, calculate evapotranspiration, and simulate groundwater recharge and baseflow.

[0130] Mesoscale model construction: collect high-resolution DEM and building data, organize urban pipe network data, and extract impervious area distribution.

[0131] Urban hydrodynamic model setting, select 2D hydrodynamic model (such as MIKE 21, HEC-RAS 2D, etc.), generate calculation grid, set boundary conditions (interfacing with macro model), surface-pipeline network coupling, build surface flow-pipeline network flow coupling model, set manhole cover overflow and backflow mechanism.

[0132] Urban hydrodynamic simulation, simulate overland flow, calculate pipe network flow and water level, identify waterlogging points and overflow areas

[0133] Micro-scale model construction, key node identification, identification of key areas of concern based on meso-scale model results, and selection of typical overflow facilities or key nodes.

[0134] High-precision data collection, high-precision topographic survey (such as laser scanning), collection of detailed structural parameters of facilities,

[0135] Refine the model construction, select a 3D CFD model (such as FLOW-3D, OpenFOAM, etc.), generate a high-precision computational grid, and set detailed boundary conditions (interfacing with the mesoscopic model).

[0136] Detailed simulation of key nodes, simulation of local water flow details, calculation of flow fields around hydraulic structures, and evaluation of facility performance and efficiency.

[0137] Model coupling and data exchange, extracting the urban area inflow boundary conditions from the macro model, using the groundwater level of the macro model as the bottom boundary of the meso model, and the outflow of the meso model as the internal source and sink of the macro model

[0138] The local water level and flow conditions of key nodes are extracted from the mesoscopic model, and the refined results of the microscopic model are fed back to the local grid of the mesoscopic model.

[0139] Multi-scale data assimilation, design a multi-scale Kalman filter framework, and combine observation data of different scales at the same time

[0140] Model operation and iteration, initialization, setting initial conditions, determining time step and simulation period,

[0141] Cascade simulation: First run the macro model with a large time step, run the meso model with multiple small time steps based on the macro results, and run the micro model with even smaller time steps for key nodes.

[0142] Results are fed back and corrected. Micro results correct local parameters of the meso model. Meso results update urban regional generalization parameters of the macro model.

[0143] Check the multiscale water balance, adjust coupling parameters if necessary and repeat.

[0144] In another embodiment of the present application, in S41, during the calculation, the region-by-region parallel calculation is adopted, including the following steps:

[0145] Initialize the computational domain, import high-resolution DEM data and city boundary data, define the computational grid resolution and coordinate system, and generate the initial computational grid;

[0146] The initial domain decomposition is performed based on the graph segmentation algorithm, the adjacency graph between grid cells is constructed, the weights between grid cells are calculated (based on elevation difference, water flow direction, etc.), and the METIS algorithm is applied to perform graph segmentation to obtain the initial subdomain.

[0147] Identify the subdomain boundary grid cells, calculate the hydraulic properties of the boundary grid (such as flow rate, water level gradient), and adjust the boundary position based on the hydraulic properties to minimize the cross-boundary flow.

[0148] Calculate the computational complexity of each subdomain (based on terrain complexity, number of grids, etc.), and adjust the subdomain size to balance the computational load of each subdomain.

[0149] Overlap regions are generated, overlapping layers are extended at subdomain boundaries, and the overlap width is determined (based on the maximum expected flow velocity and time step).

[0150] The computing framework is set up, computing resources (CPU cores or GPUs) are allocated to each subdomain, a communication mechanism between subdomains is established (such as the MPI protocol), and a data exchange strategy is designed (such as the Schwarz alternation method).

[0151] Boundary condition processing: determine external boundary conditions (such as inflow boundary and outflow boundary), set internal boundary conditions between subdomains, define overlapping area data interpolation methods (such as bilinear interpolation), and establish water balance constraints between subdomains.

[0152] Parallel solving and synchronization: Each subdomain solves the hydrodynamic equations independently, exchanges data between subdomains after a predefined time step, checks convergence, adjusts the time step if necessary, and repeats until the simulation end time is reached.

[0153] In another embodiment of the present application, the parameter optimization process of the hydrological and hydrodynamic model is as follows:

[0154] Define the optimization objective function set {f1(x), f2(x), ..., fm(x)}, select performance indicators (such as NSE, RMSE, water balance error, etc.), and construct the objective function expression.

[0155] Define the hyperparameter search space X, including determining the hyperparameters to be optimized and their value ranges, and performing necessary parameter transformations (such as logarithmic transformations).

[0156] Generate initial sampling points, use Latin hypercube sampling to generate an initial point set, and evaluate the objective function value of the initial points.

[0157] Gaussian process surrogate model construction, constructing an independent Gaussian process model for each objective function,

[0158] Select a kernel function (such as the Matérn 5 / 2 kernel) and use maximum likelihood estimation to optimize the hyperparameters of the Gaussian process.

[0159] Construct a multi-objective Gaussian process model, define the correlation structure between objectives, and realize joint prediction function

[0160] Design and optimize the acquisition function, calculate the predicted Pareto front, use the fast non-dominated sorting algorithm, calculate the hypervolume index, construct the expected improvement (EI) acquisition function, calculate the single-objective EI, aggregate the multi-objective EI (such as product or sum). Optimize the acquisition function, maximize the EI using a global optimization algorithm (such as DIRECT), and select the next evaluation point.

[0161] Run the hydrological and hydrodynamic model at the selected points, calculate the true objective function value, update the Gaussian process model, add new observation data, and re-optimize the Gaussian process hyperparameters.

[0162] Check the termination conditions (such as the maximum number of iterations, convergence criteria). If the termination conditions are not met, return to recalculate.

[0163] Output the final Pareto optimal solution set

[0164] Visualize the Pareto front and perform sensitivity analysis to select the final hyperparameter configuration.

[0165] In another embodiment of the present application, during the training process of the hydrological and hydrodynamic model, error analysis between the predicted value and the measured value is also included, and the specific process is as follows:

[0166] Obtain the error sequence and perform error decomposition as follows:

[0167] Step 1: Wavelet transform multi-scale analysis

[0168] Step 11, select a suitable wavelet basis function;

[0169] Step 111, import the PyWavelets library;

[0170] Step 112, define a candidate wavelet function list, including 'db4', 'db8', 'sym8', 'coif3';

[0171] Step 113, evaluating each candidate wavelet function;

[0172] Step 1131, performing wavelet transform on the error sequence;

[0173] Step 1132, calculating the reconstruction error;

[0174] Step 1133, calculating the frequency response characteristics;

[0175] Step 114, selecting a final wavelet function based on minimum reconstruction error and optimal frequency separation performance;

[0176] Step 12, performing discrete wavelet transform on the error sequence;

[0177] Step 121, convert the error sequence into a NumPy array;

[0178] Step 122, determining the number of decomposition layers;

[0179] Step 1221, calculate the sequence length n;

[0180] Step 1222, decomposition level number J=int(log2(n));

[0181] Step 123, use PyWavelets to perform discrete wavelet transform;

[0182] Step 1231, call pywt.wavedec function;

[0183] Step 1232, specifying the selected wavelet function and the number of decomposition levels;

[0184] Step 124, obtaining an approximation coefficient (cA) and a detail coefficient (cD);

[0185] Step 13, decomposing the error into different frequency components;

[0186] Step 131, reconstructing detail coefficients of each layer;

[0187] Step 1311, use pywt.reconstruct function;

[0188] Step 1312, setting other layer coefficients to zero;

[0189] Step 1313, obtaining the error component of the corresponding frequency band;

[0190] Step 132, reconstructing the approximate coefficients to obtain the low-frequency trend;

[0191] Step 133, calculating the energy distribution of each frequency band;

[0192] Step 1331, calculating the root mean square (RMS) for each frequency band component;

[0193] Step 1332, normalize to obtain percentage energy distribution;

[0194] S22. Statistical analysis

[0195] S221, calculate the system deviation (mean deviation);

[0196] S2211, calculating the sum of the error sequence;

[0197] S2212, calculating the length n of the error sequence;

[0198] S2213, system deviation = total / n;

[0199] S222, estimate random error variance;

[0200] S2221, subtract the system bias to obtain a debiased sequence;

[0201] S2222, calculating the sum of squares of the debiased sequence;

[0202] S2223, random error variance = sum of squares / (n-1);

[0203] S2224. Calculate the standard deviation of random error (square root of variance);

[0204] S223, identifying periodic patterns;

[0205] S2231, performing fast Fourier transform (FFT);

[0206] S22311, import numpy.fft module;

[0207] S22312, apply np.fft.fft function to the debiased sequence;

[0208] S22313, calculate frequency axis;

[0209] S2232. Calculate power spectral density (PSD);

[0210] S22321, calculating the absolute value square of the FFT result;

[0211] S22322, normalized PSD;

[0212] S2233, identify significant cycles;

[0213] S22331, set PSD threshold (mean + 2 times standard deviation);

[0214] S22332, find the frequency points exceeding the threshold;

[0215] S22333. Convert significant frequency to period (period = 1 / frequency);

[0216] S2234, use autocorrelation analysis to confirm periodicity;

[0217] S22341, calculating the autocorrelation function of the debiased sequence;

[0218] S22342, identifying the peak of the autocorrelation function;

[0219] S22343, verify the cycle of FFT recognition;

[0220] S224, Comprehensive report generation

[0221] S2241. Summarize systematic deviations, random errors, and periodicity information;

[0222] S2242, generating an error decomposition pie chart;

[0223] S22421. Calculate the proportion of systematic deviation, random error and periodic component;

[0224] S22422. Create pie charts using matplotlib;

[0225] S2243, generating a multi-scale error time series graph;

[0226] S22431. Use matplotlib to create subplots;

[0227] S22432. Draw error components of different scales in each sub-graph;

[0228] S2244, generating a periodicity analysis diagram;

[0229] S22441, draw a power spectrum density diagram;

[0230] S22442, mark significant cycles;

[0231] S22443. Draw the autocorrelation function graph.

[0232] In another embodiment of the present application, it may also be:

[0233] Step 1: Wavelet transform multi-scale analysis

[0234] Step 11: Select appropriate wavelet basis function

[0235] The selection process involves evaluating multiple wavelet families, such as Daubechies (db), Symlets (sym), and Coiflets (coif). The main considerations include orthogonality, compact support, and the order of vanishing moments. For hydrological time series, db4 or sym8 perform well. The evaluation method involves performing wavelet transforms on sample data and comparing reconstruction errors and frequency separation capabilities. It is implemented using Python's PyWavelets library, such as pywt.wavelist() to list available wavelets, pywt.wavedec() to perform decomposition, and then compare the performance of different wavelets.

[0236] Step 12: Perform discrete wavelet transform on the error sequence

[0237] First, convert the error sequence into an equally spaced NumPy array. Determine the maximum decomposition level J, usually J=int(log2(n)), where n is the length of the sequence. Use the pywt.wavedec(data,wavelet,level=J) function to perform a multi-level discrete wavelet transform. This will produce a list containing approximate coefficients (cA) and detail coefficients (cD) at multiple scales. The approximate coefficients represent the low-frequency trend of the sequence, while the detail coefficients represent high-frequency fluctuations at different scales.

[0238] Step 13: Decompose the error into different frequency components

[0239] Reconstruct the detail coefficients of each layer and the approximate coefficients of the last layer to obtain different frequency components. Use pywt.waverec([cA]+[None]*J,wavelet) to reconstruct the approximate part.

[0240] pywt.waverec([None]i+[cD[i]]+[None](Ji-1),wavelet) Reconstruct the i-th layer of details. Compute the energy distribution of each component by calculating the root mean square (RMS) of each component and normalizing it. Draw a multi-resolution analysis with time on the x-axis and different scales on the y-axis, and colors indicating the intensity of fluctuations.

[0241] Step 2: Statistical analysis

[0242] Step 21. Calculate the system deviation (mean deviation)

[0243] Systematic bias reflects the persistent deviation between the predicted value and the actual value. It is calculated as the arithmetic mean of the error sequence: bias = sum(errors) / len(errors).

[0244] Calculate the relative bias: relative_bias = bias / mean(observations). Plot the histogram and QQ plot of the error to check the distribution characteristics of the bias. If the histogram is asymmetric or the QQ plot deviates from a straight line, it may indicate nonlinear bias.

[0245] Step 22: Estimate the random error variance

[0246] First, the systematic bias is subtracted from the original error to obtain the debiased series. The random error variance is calculated as: var=sum((debiased_errors)**2) / (n-1), where n is the number of samples. The standard deviation is calculated as: std=sqrt(var). The confidence interval of the variance is estimated using the bootstrapping method, and the variance is calculated by repeated sampling to obtain the distribution of the variance. The autocorrelation plot (ACF) and partial autocorrelation plot (PACF) of the random error are plotted to check whether there is autocorrelation.

[0247] Step 23: Identify periodic patterns

[0248] Apply Fast Fourier Transform (FFT) to analyze periodicity. Use np.fft.fft() to calculate the spectrum and np.fft.fftfreq() to generate the corresponding frequency axis. Calculate the power spectral density (PSD): psd = np.abs(fft_result)**2 / len(fft_result). Identify the significant peaks in the PSD and convert the corresponding frequencies to periods. Use the Lomb-Scargle periodogram analysis method to handle non-equally spaced data. Calculate the autocorrelation function (ACF) to verify the periodicity and look for periodic peaks in the ACF. Apply continuous wavelet transform CWT to detect time-varying periodicity and visualize the time-frequency diagram to reveal the changes in the intensity of periodicity over time.

[0249] Among them, the autocorrelation function (ACF) is calculated to verify the periodicity, and the periodic peak in the ACF is found. The process of using continuous wavelet transform (CWT) to detect time-varying periodicity, that is, the process of identifying periodic patterns and using continuous wavelet transform (CWT) to detect time-varying periodicity, is specifically:

[0250] Compute the autocorrelation function (ACF) to verify periodicity:

[0251] Collect time series data and check whether the time series data is equally spaced. If not, interpolation is required. Remove the mean of the series to obtain a zero-mean series.

[0252] Calculate ACF: Use numpy's correlate function to calculate autocorrelation: acf = np.correlate(data,data, mode='full'), keep only the positive lag part: acf = acf[len(acf) / / 2:]; normalize ACF: acf / =acf[0].

[0253] Determine the lag range: usually calculated to 1 / 3 or 1 / 2 of the sequence length to balance computational efficiency and information content.

[0254] Find periodic peaks in the ACF: Set a threshold of 2 times the standard deviation of the ACF. Find local maxima that exceed the threshold. Calculate the lag difference between adjacent peaks, and these differences are possible cycles.

[0255] Significance test: Use the Ljung-Box test or Box-Pierce test to assess the statistical significance of the ACF peak. Calculate the confidence interval, usually using ±2 / sqrt(n) as the 95% confidence interval, where n is the sequence length.

[0256] Visualization: Draw an ACF plot with lag on the x-axis and correlation coefficient on the y-axis. Mark significant periodic peaks and confidence intervals on the plot.

[0257] Applying the Continuous Wavelet Transform (CWT) to detect time-varying periodicity:

[0258] Select the wavelet function: For periodic analysis, choose the Morlet wavelet because it has good localization properties in both time and frequency domains. Define the Morlet wavelet parameters, such as the center frequency ω0 (usually set to 6).

[0259] Determine the scale range: Convert the period range of interest to the corresponding scale. The scale is usually logarithmically spaced, such as:

[0260] scales = np.logspace(np.log10(min_scale), np.log10(max_scale), num_scales).

[0261] Execute CWT: Use the cwt function of the PyWavelets library: coefficients, frequencies = pywt.cwt(data, scales, 'morl', sampling_period=1 / fs), where fs is the sampling frequency.

[0262] Calculate the wavelet power spectrum: Power spectrum = |coefficients| 2 ;

[0263] Time-varying periodicity analysis: Identify high-power regions on the time-scale plane. For each time point, find the scale with the highest power and convert it to the corresponding period. Track the changes in the high-power regions and analyze how the periodicity changes over time.

[0264] Statistical significance testing: Generate a red noise background spectrum. Calculate the significance level at each time-scale point. Estimate confidence intervals using Monte Carlo methods.

[0265] Visualization: Plot a heat map of the wavelet power spectrum with time on the x-axis and period (or frequency) on the y-axis. Use a color map to represent power intensity. Mark statistically significant regions on the plot. Add a cone of influence to indicate edge effects.

[0266] Extract the main periodic components: Time average the power spectrum to obtain the global wavelet spectrum. Identify the main periods in the global spectrum. Reconstruct these main periods to obtain their contribution to the original signal.

[0267] Comprehensive analysis: Compare the results of ACF and CWT to verify the periodicity found. Analyze how the intensity and phase of the periodicity change over time. Combined with the problem context, explain the physical significance of the periodicity found.

[0268] It should be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

Claims

1. A method for constructing sponge city overflow facilities based on high-precision rain and flood simulation, characterized in that: The steps include: Step S1, determine the scope of the urban area, collect basic data and pre-process; Step S2: based on the preprocessed basic data, extract the hydrological and meteorological data, call the pre-built rain and flood simulation model to analyze the waterlogging situation, and preliminarily extract and screen the overflow area according to the waterlogging situation to construct an overflow area set; Step S3: based on the preprocessed basic data, extract the ecological pedestrian channel and construct the existing ecological and pedestrian orientation map; for each overflow area, extract the newly added ecological and pedestrian orientation sub-map, and superimpose it with the existing ecological and pedestrian orientation map to form an overflow area set with connectivity information, and generate at least two overflow area planning schemes; Step S4: for each overflow area planning scheme in the overflow area set with connectivity information, a rainstorm simulation model is used to perform a refined simulation, and overflow facility parameters under typical rainstorm conditions are given to form an overflow facility parameter set; The step S3 is further as follows: Step S31, extracting the urban ecological network and the urban pedestrian network respectively based on the preprocessed basic data; Step S32: read the urban ecological network and the urban pedestrian network and analyze the connectivity to obtain the existing ecological and pedestrian maps; Step S33: for each overflow area, extract green space, water system, road and trail respectively, construct a new ecological sub-graph and a new walking sub-graph, analyze the connectivity, obtain the new ecological and walking graphs, and make topological association with the existing ecological and walking graphs, and extract the topological association information; Step S34: adding the topological association information to the overflow area set to form an overflow area set containing ecological and pedestrian connectivity information; In step S33, the process of obtaining a newly added ecological and walking graph and topologically associating it with the existing ecological and walking graph is specifically as follows: Step S331: abstract the overflow area set into an undirected graph, and calculate the spatial distance matrix between the overflow areas based on the undirected graph and use it as the edge weight; Step S332: Use the Louvain algorithm to perform community detection on the graph G, identify closely connected overflow area communities, and form an overflow area community set; for each overflow area community, construct an overflow area subgraph within it; Step S333: for each overflow area community, superimpose its corresponding overflow area sub-graph and the influence parameter to form a blue-green space network at the community scale; Step S334: On the blue-green space network at the community scale, the minimum cumulative resistance model is used to calculate the connectivity matrix, and the improved Kruskal algorithm is used to extract the minimum ecological corridor to form an ecological orientation map at the community scale; and combined with the road network data and the residents' outdoor activity preference data, the weighted Voronoi partition is used to generate a pedestrian walking orientation map that takes into account ecological connectivity and accessibility, and a new ecological and walking map is obtained; Step S335: splice and optimize the existing ecological and pedestrian map and the newly added ecological and pedestrian map to form an overall urban-scale ecological and pedestrian map.

2. The method for constructing a sponge city overflow facility based on high-precision rain and flood simulation according to claim 1, characterized in that: The step S1 is specifically as follows: Step S11, determine the scope of the urban area and collect basic data, the basic data includes urban DEM data, urban basic information, hydrological and meteorological data and ecological environment data; urban basic information includes land use status map and impervious surface distribution data; Step S12: convert the format of the basic data, unify the coordinate system and spatial resolution, obtain the processed basic data and store it; Step S13: construct a three-dimensional terrain model based on the processed basic data to provide terrain boundary conditions for stormwater simulation.

3. The method for constructing a sponge city overflow facility based on high-precision rain and flood simulation according to claim 2, characterized in that: The step S2 is further as follows: Step S21, extracting the pre-processed hydrological and meteorological data to generate rainfall scenario data with different return periods; Step S22: for each rainfall scenario data, call the pre-built rain and flood simulation model to preliminarily simulate the generation and convergence process of surface runoff, as well as the flow process of rainwater on the surface and in the pipe network, and analyze the depth and range of water accumulation in each area; Step S23: extracting areas where the water depth exceeds the threshold as potential overflow areas, and screening and merging the potential overflow areas in combination with terrain and underlying surface factors to obtain a preliminary overflow area set.

4. The method for constructing a sponge city overflow facility based on high-precision rain and flood simulation according to claim 3 is characterized in that: The step S4 is further as follows: Step S41: calling the rain and flood simulation model, performing a refined simulation for each overflow area planning scheme, and obtaining the water depth, range and duration of the overflow area under each rainfall scenario; Identify key water accumulation nodes and confluence paths in the overflow area; Step S42: Optimize the design parameters of the overflow facilities for the key water accumulation nodes of each overflow area under each rain and flood situation to obtain a parameter combination; Step S43: using a risk analysis method to evaluate the peak shaving and flood reduction effect and failure risk of the overflow facility under each parameter combination; Generate an optimal overflow facility parameter set for each overflow area in each overflow area planning scheme, and attach corresponding risk assessment information; Step S44: construct an overflow area planning scheme evaluation model, evaluate the optimal overflow facility parameter set, and obtain a facility parameter set that meets preset conditions as the overflow setting parameter set.

5. The method for constructing a sponge city overflow facility based on high-precision rain and flood simulation according to claim 4, characterized in that: In step S32, the process of reading the urban ecological network and the urban pedestrian network and analyzing the connectivity to obtain the existing ecological map is specifically as follows: Step S321, obtaining pre-processed basic data, obtaining influencing parameters including terrain undulation, land use type, traffic roads and habitat quality, and constructing the resistance surface of the urban blue-green space network based on the influencing parameters; Step S322: Taking each blue-green space patch as a source, calculate the minimum cumulative resistance distance from it to other blue-green space patches to generate a cost distance matrix; calculate the connectivity index of the blue-green space network based on the cost distance matrix, obtain the connectivity matrix and analyze the vulnerability of the network connectivity; the connectivity index includes the overall connectivity and the key patch connectivity; Step S323: Based on the connectivity matrix, the Kruskal algorithm is used to iteratively select the edge with the minimum cost, generate the minimum ecological corridor and perform conflict detection to obtain the minimum ecological corridor network; the nodes of the minimum ecological corridor network are used as seed points, and the weighted Voronoi diagram is generated in combination with the road network data; Step S324: Calculate the ecological connectivity and accessibility between each Voronoi cell and adjacent Voronoi cells in the weighted Voronoi diagram.

6. The method for constructing a sponge city overflow facility based on high-precision rain and flood simulation according to claim 4, characterized in that: Step S41 also includes evaluating the importance of the overflow area, specifically: Step S411, extracting an ecological orientation map, and extracting an ecological attribute feature vector for each node in the ecological orientation map, wherein the node includes an overflow area; the ecological attribute feature vector includes a green area, a water area, and a biodiversity index; Step S412, calculating the connectivity and betweenness centrality of the nodes, and using the ecological flow potential evaluation model, using the weighted geometric mean method to integrate the ecological attributes, connectivity and betweenness centrality of the nodes to obtain the ecological flow potential score of each node; Step S413, extracting a pedestrian walking guide map, and extracting outdoor activity preference feature vectors including visit frequency, stay time, and activity type diversity from each node in the pedestrian walking guide map; Step S414: Calculate the connectivity and load capacity of the node, adopt the walking gravitational potential evaluation model, and use Choquet integration to perform nonlinear fusion of the node's outdoor activity preference, connectivity, and load capacity to obtain a walking gravitational potential score for each node; Step S415, construct a fuzzy decision matrix, where the rows represent overflow areas, the columns represent two decision criteria, namely, ecological flow potential and spreading gravitational potential, and the matrix elements are interval fuzzy numbers; Considering the relative importance of ecological benefits and pedestrian activity benefits, the weight vector of the decision criterion is determined; Step S416, calculate the distance of each overflow area relative to the positive ideal solution and the negative ideal solution, and use the Hamming distance to measure the difference between interval fuzzy numbers; calculate the comprehensive score of each overflow area, weighing its superiority relative to the positive ideal solution and its inferiority relative to the negative ideal solution; sort the overflow areas according to the comprehensive score to obtain the overflow area importance parameter.

7. The method for constructing a sponge city overflow facility based on high-precision rain and flood simulation according to claim 4, characterized in that: In step S23, the process of screening and merging potential overflow areas also includes optimizing the screening of overflow areas, specifically: Step S231, constructing an overflow area planning scheme evaluation model, wherein the model optimization objectives include peak shaving and flood reduction effects, ecological connectivity, and engineering costs; Step S232: using a genetic algorithm to solve the optimization model and obtain a Pareto optimal solution set for the overflow facility layout; Step S233, evaluating the ecological connectivity improvement effect of overflow facility layout under different Pareto optimal solutions, and selecting the layout plan with the best ecological benefit; Step S234: When in use, make a decision on the overflow facility layout plan, use the hierarchical analysis method to determine the weights of different factors, and sort and optimize the alternative layout plans.

8. A sponge city overflow facility construction system based on high-precision rain and flood simulation, characterized in that: include: at least one processor; as well as, a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the method for constructing sponge city overflow facilities based on high-precision rain and flood simulation as described in any one of claims 1 to 7.