Multi-objective rainwater pipe network flow calculation optimization method and system based on SWMM

Abstract

The application discloses a multi-target rainwater pipe network flow calculation optimization method and system based on SWMM, relates to the technical field of hydraulic calculation, and comprises the following steps: candidate pipe network diagrams are constructed and initial pipe network topologies are generated, hydraulic simulation is carried out under a multi-rain type scenario to obtain hydraulic time sequence state data; based on the data, confluence response segments are extracted and confluence competition relations are constructed, key scenarios and key time windows are identified, and a competition constraint flow sequence is formed; according to this, pipe section capacity constraints are determined, node competition risk weights are calculated, and candidate edge weights are updated; based on the updated edge weights, pipe network topology optimization is executed, multi-target evaluation is carried out in combination with pipe network cost and expected annual loss, and the optimized pipe section flow calculation result is output. The operation reliability of the rainwater pipe network under complex rainfall conditions is improved.

Description

Technical Field

[0001] This invention relates to the field of hydraulic calculation technology, and in particular to a method and system for multi-objective stormwater network flow calculation and optimization based on SWMM. Background Technology

[0002] With the acceleration of urbanization, the impervious area of ​​cities is constantly increasing, and the runoff of rainwater is significantly increasing. Under heavy rainfall conditions, rainwater pipe networks are prone to localized congestion and overflow at nodes. Current rainwater pipe network design and analysis typically rely on hydraulic simulation tools such as SWMM to calculate and analyze the network's operational status under different rainfall scenarios. However, existing methods are mostly based on single rainfall scenarios or static load distribution for pipe network layout and capacity verification, making it difficult to reflect the dynamic competition process caused by the superposition of flow from multiple sub-catchments at the same node.

[0003] In actual operation, the runoff in different sub-catchments varies over time. Changes in arrival time, peak overlap, and duration can lead to significant fluctuations in instantaneous flow at nodes, further affecting the water conveyance capacity allocation of downstream pipe sections. Existing methods typically do not provide a detailed characterization of the temporal overlap and capacity occupancy relationships in the aforementioned runoff processes. This makes it difficult to accurately identify key constraint locations in rainy scenarios, thereby affecting the rationality and safety margin of the pipeline network design. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a multi-objective stormwater network flow calculation optimization method based on SWMM to solve the problem that existing stormwater network flow calculation methods are difficult to reflect the multi-source confluence competition process and its impact on capacity constraints, resulting in the difficulty of balancing drainage safety and economy in the network optimization results.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a multi-objective stormwater pipe network flow calculation optimization method based on SWMM, which includes acquiring multi-source drainage basic data, performing spatial alignment and attribute extraction, and constructing a candidate pipe network map;

[0008] An initial pipeline topology is generated based on the candidate pipeline diagram, and the initial pipeline topology is input into the SWMM model. Hydraulic simulation is performed under a rainy scenario to obtain the hydraulic time series state data of nodes and pipe segments.

[0009] Based on the hydraulic time-series state data, the confluence response segments of each sub-catchment area to the candidate nodes under different rainfall scenarios are extracted, and the confluence competition relationship is constructed according to the arrival time relationship, overlap relationship and capacity occupancy relationship between the confluence response segments;

[0010] Based on the aforementioned confluence competition relationship, key scenarios and key time windows are identified, the flow rate of each pipe segment in the competition constraint interval is determined, and the capacity constraint is determined based on the flow rate in the competition constraint interval.

[0011] Combining the aforementioned confluence competition relationship, the flow rate within the competition constraint interval, and historical water accumulation information, the node competition risk weight is calculated, and the node competition risk weight is written back to the candidate edges flowing into the corresponding nodes to form updated edge weights.

[0012] The pipeline topology is regenerated based on the updated edge weights, and a multi-objective evaluation index including pipeline cost index and expected annual loss index is constructed. The topology optimization iteration is performed based on the multi-objective evaluation index until the multi-objective convergence condition is met, and the optimized pipeline flow calculation results are output.

[0013] As a preferred embodiment of the multi-objective stormwater pipe network flow calculation and optimization method based on SWMM described in this invention, the multi-source drainage basic data includes road centerline data, terrain elevation data, plot boundary data, land use data, and historical water accumulation data.

[0014] The construction of the candidate pipeline network map includes: dividing the sub-catchments based on the topographic elevation data; extracting the area, average slope, and impermeability of each sub-catchment; calculating the runoff demand index based on the area, slope, and impermeability of the sub-catchments; mapping each sub-catchment to the corresponding road nodes to form a runoff relationship; constructing a candidate pipeline network map using road intersections and outlets as nodes and road centerline segments as candidate edges; constructing basic edge weights based on the length, elevation difference, historical water accumulation sensitivity, and runoff demand index differences of connected nodes of the candidate edges; and performing directional processing on the candidate edges based on the node elevation relationship.

[0015] As a preferred embodiment of the multi-objective stormwater pipe network flow calculation optimization method based on SWMM described in this invention, the hydraulic simulation under the rainy scenario includes: performing path search based on the basic edge weights of candidate edges to generate an initial pipe network topology; mapping the nodes in the initial pipe network topology to node objects in the SWMM model, mapping the candidate edges to pipe segment objects in the SWMM model, mapping the sub-catchment areas to sub-catchment area objects in the SWMM model, configuring hydrological parameters based on the area and impermeability of the sub-catchment areas, constructing a set of rainy scenarios containing different rainfall durations and peak locations, and inputting the rainy scenarios into the SWMM model; performing hydraulic simulation on the SWMM model based on the rainy scenarios, obtaining the flow time series data, overflow time series data of each node, and full flow state data of each pipe segment, and forming a unified hydraulic time series state data set.

[0016] As a preferred embodiment of the multi-objective stormwater network flow calculation and optimization method based on SWMM described in this invention, the construction of the confluence competition relationship includes: based on the hydraulic time series state data set, extracting the flow response process of each sub-catchment to each candidate node under different rainfall scenarios according to a unified time step; for each sub-catchment at each candidate node, identifying the flow rise time, peak time and fall end time, and forming the corresponding confluence response segment;

[0017] For the confluence response segments of different sub-catchments flowing toward the same candidate node, the arrival time difference, overlap duration, and composite flow occupancy during the overlap period are statistically analyzed between the segments. Based on the arrival time difference, overlap duration, and composite flow occupancy, the confluence competition intensity of different sub-catchments at the corresponding candidate node is determined. The confluence competition intensity at each candidate node under each rainfall scenario is summarized to form a confluence competition relationship that reflects the competition relationship between sub-catchments and the degree of competition between candidate nodes.

[0018] As a preferred embodiment of the multi-objective stormwater pipe network flow calculation optimization method based on SWMM described in this invention, the identification of key scenarios and key time windows includes: based on the confluence competition relationship, respectively calculating the cumulative value of competition intensity of each candidate node under different rainfall scenarios, and determining the rainfall scenario in which the cumulative value of competition intensity meets the preset conditions as the key scenario; for the key scenario, extracting the time segment in which the competition intensity at each candidate node continuously increases and remains above the preset threshold as the key time window;

[0019] The sub-catchments and corresponding candidate nodes participating in the competition within each key time window are associated to form competition units under key scenarios; based on the pipe segment flow time sequence corresponding to each competition unit, pipe segment flow segments corresponding to the key time windows are extracted; based on the pipe segment flow segments and the corresponding competition intensity, a competition constraint flow sequence characterizing the multi-source confluence competition process is formed.

[0020] As a preferred embodiment of the multi-objective stormwater pipe network flow calculation and optimization method based on SWMM described in this invention, the determination of the competitive constraint interval flow includes: determining the basic flow demand of each pipe segment under stable transport conditions and the peak transport demand under competitive superposition conditions based on the competitive constraint flow sequences corresponding to each key scenario and each key time window; determining the competitive constraint interval flow of each pipe segment based on the basic flow demand and the peak transport demand; and determining the capacity constraint of each pipe segment based on the relationship between the competitive constraint interval flow and the current transport capacity of each pipe segment.

[0021] The computational node competition risk weights include: based on the confluence competition relationship, calculating the cumulative competition intensity of each node under different key scenarios; based on the capacity constraints of the pipe segments connected to each node, determining the capacity competition mismatch degree corresponding to each node; combining the historical water accumulation information, determining the water accumulation sensitivity of each node; calculating the competition risk weight of each node based on the cumulative competition intensity, capacity competition mismatch degree, and water accumulation sensitivity; and mapping the node competition risk weights to the candidate edges flowing into the corresponding nodes to increase the edge weights of the candidate edges corresponding to high competition risk areas, forming updated edge weights.

[0022] As a preferred embodiment of the multi-objective stormwater pipe network flow calculation and optimization method based on SWMM described in this invention, the step of performing topology optimization iteration based on the multi-objective evaluation index includes: generating candidate pipe network topologies based on the updated edge weights, and forming candidate pipe network design schemes by combining optional pipe diameter combinations;

[0023] Using the pipeline cost index and the expected annual loss index as dual optimization objectives, the merits of each candidate pipeline design scheme are determined, and candidate schemes with different trade-off characteristics are retained based on their relative positions in the pipeline cost index and the expected annual loss index. Based on the merits determination results, candidate schemes that are not simultaneously degraded by other candidate schemes are retained, and a new round of candidate schemes is generated based on the retained schemes. Hydraulic simulation and multi-objective evaluation index construction are repeatedly performed on the updated candidate schemes. When the iteration meets the preset multi-objective convergence condition, the Pareto front solution set composed of the non-dominated solution set formed by the merits determination results is output. Based on the distribution relationship of each solution in the Pareto front solution set between the pipeline cost index and the expected annual loss index, the pipeline topology corresponding to the representative optimal solution is selected, and the corresponding pipe segment flow calculation results are output.

[0024] Secondly, this invention provides a multi-objective stormwater pipe network flow calculation and optimization system based on SWMM, comprising: a data processing module for acquiring multi-source drainage basic data and constructing candidate pipe network diagrams; a simulation module for generating an initial pipe network topology and performing SWMM hydraulic simulation to obtain hydraulic time-series state data; a competition analysis module for constructing confluence competition relationships and determining flow and capacity constraints within competition constraint intervals; an edge weight update module for calculating node competition risk weights and updating the edge weights of candidate edges; and an optimization module for generating a pipe network topology based on updated edge weights, constructing multi-objective evaluation indicators, performing topology optimization iterations, and outputting pipe segment flow calculation results.

[0025] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the multi-objective stormwater network flow calculation and optimization method based on SWMM as described in the first aspect of the present invention.

[0026] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the multi-objective stormwater network flow calculation and optimization method based on SWMM as described in the first aspect of the present invention.

[0027] The beneficial effects of this invention are as follows: By uniformly processing hydraulic time-series state data under rainy scenarios, the confluence response segments of each sub-catchment at candidate nodes are extracted, and confluence competition relationships are constructed based on arrival time relationships, overlap relationships, and capacity occupancy relationships. This enables the identification of key scenarios and key time windows that exert actual constraints on nodes and pipe segments during the superposition of multiple confluences. Furthermore, the basic flow demand and peak transport demand of pipe segments are determined through competitive constraint flow sequences, and node competition risk weights are calculated by combining historical water accumulation information. Risk information is then written back to candidate edges to adjust edge weights, thereby guiding network topology reconfiguration. Through the above processing, the network optimization process is transformed from traditional static load allocation to a constraint-driven approach based on dynamic confluence competition. Furthermore, multi-objective optimization is performed by combining cost indicators and expected annual loss indicators, thereby improving the accuracy of flow calculation while achieving synergistic optimization of drainage safety and economy, and enhancing the operational reliability of stormwater networks under complex rainfall conditions. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of a multi-objective stormwater network flow calculation optimization method based on SWMM.

[0030] Figure 2 A schematic diagram illustrating the confluence competition relationship in the multi-objective stormwater pipe network flow calculation optimization method based on SWMM.

[0031] Figure 3 This is a computer equipment diagram for a multi-objective stormwater network flow calculation and optimization method based on SWMM. Detailed Implementation

[0032] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0033] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0034] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0035] Reference Figures 1-3 This is one embodiment of the present invention, which provides a multi-objective stormwater network flow calculation optimization method based on SWMM, including the following steps:

[0036] S1: Obtain basic data of multi-source drainage, perform spatial alignment and attribute extraction, and construct candidate pipe network diagrams; generate initial pipe network topology based on candidate pipe network diagrams, and input the initial pipe network topology into the SWMM model to perform hydraulic simulation under rainy conditions, and obtain hydraulic time series state data of nodes and pipe segments.

[0037] Acquire multi-source drainage basic data, including road centerline data, topographic elevation data, plot boundary data, land use data, and historical water accumulation data. Perform unified coordinate reference system transformation and spatial registration processing on the above data to establish correspondence among various data types within the same spatial framework.

[0038] After spatial alignment, sub-catchments were divided based on plot boundary data and topographic elevation data. For each sub-catchment, area, average slope, and impermeability attributes were extracted. The area was calculated from the sub-catchment boundaries, the average slope was obtained through elevation change statistics, and the impermeability was determined by the proportion of impermeable coverage area in the land use data.

[0039] Subsequently, each sub-catchment is mapped to its corresponding road node, forming a node confluence relationship. Specifically, using road intersections and discharge outlets as candidate nodes, sub-catchments are mapped to corresponding nodes according to spatial proximity or drainage direction, and the confluence load of multiple sub-catchments mapped to the same node is accumulated to obtain the node's comprehensive confluence demand attribute.

[0040] After establishing the nodes and their confluence attributes, a candidate pipeline network diagram is constructed using road intersections and discharge outlets as nodes and road centerline segments as candidate edges. Each candidate edge is associated with a length attribute, an elevation difference attribute, a historical waterlogging sensitivity attribute, and a difference in confluence demand between connected nodes. The length is determined by the road's geometric length, the elevation difference is determined by the elevation difference between the two endpoints, the historical waterlogging sensitivity is obtained from historical waterlogging data, and the difference in confluence demand between nodes is determined by the difference in confluence demand attribute values ​​between connected nodes.

[0041] To uniformly represent the comprehensive cost of candidate edges, a basic edge weight is calculated for each candidate edge, as follows:

[0042]

[0043] in, Indicates the connection node With nodes The basic edge weights of the candidate edges; Indicates the connection node With nodes The length of the candidate edge; Represents a node Elevation value; Represents a node Elevation value; Indicates the connection node With nodes Historical water accumulation sensitivity indicators for the regions corresponding to candidate edges; Represents a node The comprehensive convergence demand attribute value; Represents a node The comprehensive convergence demand attribute value; Represents the weighting coefficient of the length factor; The weighting coefficient of the elevation difference factor The weighting coefficients represent the historical waterlogging sensitivity factors; This represents the weighting coefficient of the node convergence demand difference factor; The normalized baseline value representing the candidate edge length; The normalized baseline value representing the elevation difference between nodes; This represents the normalized baseline value representing the difference in the overall convergence demand of nodes.

[0044] After the basic edge weights are calculated, candidate edges are oriented based on node elevation relationships. Specifically, nodes with higher elevations are directed to nodes with lower elevations to form candidate flow directions consistent with actual surface drainage trends. When the elevation difference between the two endpoints is below a preset threshold, the undetermined flow direction attribute of the candidate edge is retained, and the final flow direction is determined in subsequent initial pipeline topology generation by combining the outlet direction and global connectivity. This process ensures that the candidate pipeline map maintains the integrity of the road network while satisfying the basic physical law of drainage system flow from high to low elevations.

[0045] After the candidate pipeline network graph is constructed, an initial pipeline network topology is generated based on the basic edge weights of the candidate edges using a shortest path search strategy. Specifically, with the discharge outlet node as the confluence endpoint, a path search is performed from each access node to the discharge outlet node. The connected path with the smallest cumulative edge weight is selected as the drainage path corresponding to that access node. This process is repeated for all access nodes to obtain the initial pipeline network topology covering the study area. In this way, the initial pipeline network topology satisfies drainage direction constraints and network connectivity requirements while also considering path length, elevation adaptability, water accumulation sensitivity, and the matching relationship between node confluence requirements.

[0046] After obtaining the initial pipeline topology, it is input into the SWMM model. Specifically, nodes in the initial pipeline topology are mapped to node objects in the SWMM model, candidate edges in the initial pipeline topology are mapped to pipe segment objects in the SWMM model, and the aforementioned sub-catchments are mapped to sub-catchment objects in the SWMM model. For each sub-catchment, corresponding hydrological parameters are configured based on its area and impermeability; for each pipe segment, corresponding topological and structural parameters are configured based on its length, starting node, ending node, and node elevation relationships; for each node, corresponding node parameters are configured based on its spatial location, elevation, and confluence connection relationships. Through the above mapping, the initial pipeline topology generated from the candidate pipeline map can be directly called by the SWMM model.

[0047] After completing the SWMM model construction, a set of multiple rainfall scenarios was further constructed. This set includes scenarios with different rainfall durations and scenarios with different peak locations. Scenario scenarios with different rainfall durations are used to simulate the different impacts of short-duration heavy rainfall, moderate-duration rainfall, and long-duration continuous rainfall on the pipe network system; scenarios with different peak locations are used to simulate the different impacts on the node confluence process and pipe segment water conveyance process when the rainfall peak is located at the beginning, middle, or end of the rainfall process. Each multiple rainfall scenario is sequentially input into the SWMM model, and hydraulic simulation is performed on the initial pipe network topology.

[0048] During the hydraulic simulation, flow time-series data and overflow time-series data for each node, as well as flow time-series data and full-flow state data for each pipe segment, are acquired. All data types are then organized according to a unified time step to form a unified hydraulic time-series state data set. Specifically, the node flow time-series data characterizes the flow change process at each node throughout the simulation; the node overflow time-series data characterizes the overflow response process after a node exceeds its carrying capacity; the pipe segment flow time-series data characterizes the water conveyance change process of each pipe segment at different time steps; and the pipe segment full-flow state data characterizes whether each pipe segment reaches or approaches full-flow state at different time steps.

[0049] By performing unified coordinate transformation and spatial registration on road centerlines, topographic elevations, plot boundaries, land use, and historical water accumulation information, a stable correspondence is established between data from different sources within the same spatial framework. Compared to modeling based on a single data source, this approach avoids misjudgments of confluence paths caused by data misalignment, providing a clear spatial basis for subsequent sub-catchment division and node mapping.

[0050] In the process of sub-catchment delineation and attribute extraction, key parameters such as area, slope, and impermeability are extracted through joint constraints of topographic elevation and plot boundaries. Furthermore, the sub-catchment is mapped to road nodes based on spatial proximity and drainage direction. Compared to the traditional method of delineating catchment areas based on experience, this approach ensures that the runoff path aligns with actual topographic conditions, thereby improving the realism of the node runoff load.

[0051] In the process of constructing candidate pipeline network diagrams and generating initial topology, basic edge weights are formed by comprehensively considering side lengths, elevation differences, historical water accumulation sensitivity, and differences in node confluence requirements. Furthermore, flow direction is constrained by elevation relationships, ensuring that the generated paths satisfy both drainage direction and water conveyance load distribution. Compared to single shortest path or minimum slope path generation methods, this multi-factor constraint approach effectively reduces the risk of localized concentrated confluence.

[0052] S2: Based on the hydraulic time-series state data, extract the confluence response segments of each sub-catchment area to the candidate nodes under different rainfall scenarios, and construct confluence competition relationships according to the arrival time relationship, overlap relationship and capacity occupancy relationship between the confluence response segments; identify key scenarios and key time windows based on the confluence competition relationships, determine the flow rate of the competition constraint interval of each pipe segment, and determine the capacity constraint according to the flow rate of the competition constraint interval.

[0053] After obtaining the hydraulic time-series state data set, the flow response process of each sub-catchment to each candidate node under different rainfall scenarios is extracted according to a unified time step. Specifically, based on the flow transfer relationship between sub-catchments and candidate nodes in the initial pipeline topology, the flow contribution formed by each sub-catchment at each candidate node is read in time series to obtain the corresponding flow response process. The flow response process refers to the time-varying inflow response sequence formed by a sub-catchment to a candidate node under a certain rainfall scenario.

[0054] After obtaining the flow response process for each sub-catchment at each candidate node, the flow initiation time, peak time, and decline termination time are identified to form corresponding confluence response segments. The flow initiation time refers to the starting moment when the flow response process continuously increases relative to the initial stable flow and reaches a preset increase; the peak time refers to the moment when the flow reaches its maximum value within the confluence response segment; the decline termination time refers to the moment when the flow decreases and recovers to near the initial stable flow, and no further significant increases occur subsequently. The initial stable flow is determined by the average inflow to the node over multiple consecutive time steps before the start of rainfall; the continuous increase is judged by the positive flow increments over multiple adjacent time steps; the continuous significant decline is judged by the flow resuming continuous growth over multiple subsequent adjacent time steps and the cumulative increase exceeding a preset proportion. Through the above processing, the complete flow response process is converted into confluence response segments with clearly defined boundaries.

[0055] After forming the confluence response segments, for confluence response segments from different sub-catchments flowing towards the same candidate node, the arrival time difference, overlap duration, and composite flow occupancy during the overlap period are statistically analyzed. The arrival time difference is determined by the difference in the flow initiation time of the two confluence response segments; the overlap duration is determined by the overlap length of the time intervals of the two confluence response segments; and the composite flow occupancy is determined by the average occupancy ratio of the sum of the corresponding flows of the two sub-catchments during the overlap period relative to the transport capacity of the downstream main control pipe section corresponding to the candidate node. The downstream main control pipe section refers to the pipe section connected to the candidate node that plays a major role in drainage transport under the current rainy scenario, with priority given to the pipe section with the largest cumulative water transport volume during the critical period; when the cumulative water transport volumes are similar, the pipe section with the longer full-flow duration is selected.

[0056] To uniformly characterize the degree of competition among different sub-catchments at the same candidate node, the confluence competition intensity of each pair of sub-catchments at the candidate node is calculated as follows:

[0057]

[0058] in, Indicates the sub-watershed area Yuzihui Water Area In candidate nodes The intensity of competition at the confluence point; Indicates the sub-watershed area Yuzihui Water Area For candidate nodes The time difference of arrival between the bus response segments; Indicates the sub-watershed area Yuzihui Water Area For candidate nodes The duration of overlap between the confluence response segments; Indicates the sub-catchment area Yuzihui Water Area In candidate nodes The extent of composite traffic occupancy during overlapping time periods; Indicates candidate nodes Reference timescale under the current rainy scenario; The weighting coefficients of the arrival time difference factor; The weighting coefficients of the overlap duration factor; This represents the weighting coefficient of the synthetic flow occupancy factor. The reference time scale is determined by the span of the inflection point of all confluence response segments corresponding to the candidate nodes under the current rainy scenario; when the arrival time difference is greater than the reference time scale, the corresponding time proximity term is set to zero. Through the above processing, the arrival time relationship, overlap relationship, and capacity occupancy relationship can be uniformly represented in the same index.

[0059] After obtaining the flow competition intensity of each pair of sub-catchments at each candidate node, the flow competition intensity at each candidate node under the same rainfall scenario is summarized to form a flow competition relationship reflecting the competition relationship between sub-catchments and the degree of competition among candidate nodes. For example... Figure 2 As shown, the flow responses of different sub-catchments at the candidate node overlap in time, forming a flow competition period, and have a cumulative impact on the water conveyance capacity of the downstream pipeline during the overlap process. Here, tA, tB, and tC represent the starting times of the flow response processes formed by catchment areas A, B, and C at the candidate node, respectively, used to characterize the temporal relationship of the flow arrival from different sub-catchments.

[0060] Specifically, the competition intensity of each pair of sub-catchments flowing towards the same candidate node is summed to obtain the cumulative competition intensity value of the candidate node under the current rainy scenario. The larger the cumulative competition intensity value, the more significant the impact of multi-source confluence on the candidate node under the current rainy scenario, and the more likely it is to form a local water conveyance restriction state.

[0061] After a confluence competition relationship is established, key scenarios and key time windows are identified based on this relationship. Specifically, the cumulative competition intensity of each candidate node under different rainfall scenarios is statistically analyzed, and the nodes are sorted from highest to lowest cumulative competition intensity. Rainfall scenarios whose cumulative competition intensity meets preset conditions are identified as key scenarios. The preset conditions are preferably the set of scenarios ranking at the top of the cumulative competition intensity ranking for all rainfall scenarios, or the set of scenarios whose cumulative competition intensity is higher than the average level of all rainfall scenarios and exceeds a preset dispersion level. This setting aims to select representative scenarios from all rainfall scenarios that are more likely to trigger node competition superposition and pipeline water conveyance restriction.

[0062] For the aforementioned critical scenario, time segments where the competition intensity at each candidate node continuously increases and remains above a preset threshold are further extracted as critical time windows. The continuous increase in competition intensity means that the increments of competition intensity at multiple adjacent time steps are all positive, and the cumulative increase reaches a preset proportion of the average competition intensity of the current candidate node under the current critical scenario. Maintaining above the preset threshold means that the competition intensity at each time step within the corresponding time segment is not lower than the high-order distribution threshold of the competition intensity time series of the candidate node under the current critical scenario. Through the above processing, the critical time windows accurately correspond to time segments where competition is significantly enhanced and persistent, rather than just a single instantaneous peak moment.

[0063] After identifying key scenarios and key time windows, the competing sub-catchments and corresponding candidate nodes within each key time window are associated to form competing units under the key scenarios. A competing unit refers to a set of sub-catchments and their corresponding downstream main control pipe segments that compete for the same candidate node within the same key scenario and key time window. Subsequently, based on the pipe segment flow time series corresponding to each competing unit, pipe segment flow segments corresponding to the key time windows are extracted. Based on these pipe segment flow segments and their corresponding competition intensities, a competition constraint flow sequence characterizing the multi-source confluence competition process is formed. This competition constraint flow sequence serves as the input for subsequently determining the flow and capacity constraints of each pipe segment's competition constraint interval.

[0064] In the stage of utilizing hydraulic simulation results, different rainfall durations and peak locations are combined to form a multi-rainfall scenario, and hydraulic time-series state data of nodes and pipe sections are extracted, so that the operating characteristics of the pipeline network can be fully reflected under various operating conditions. Compared with the method of selecting only typical rainfall for analysis, multi-scenario input can cover different extreme cases, thereby improving the representativeness of subsequent analysis results.

[0065] During the flow response process identification, the flow change process of candidate nodes in sub-catchments is extracted by using a unified time step, and the boundaries of rise, peak, and fall are identified, transforming continuous time-series data into response segments with clear boundaries. Compared with direct analysis using complete flow curves, this segmentation process can highlight key confluence stages and reduce the interference of invalid fluctuations on the analysis results.

[0066] In constructing the confluence competition relationship, a comprehensive analysis of the arrival time difference, overlap duration, and capacity occupancy between response segments of different sub-catchments allows for the quantification of the mutual influence relationships between multi-source confluences. Compared to judging load solely based on peak superposition, this approach reflects the superposition process over time, thus more accurately depicting the true stress state at the nodes.

[0067] S3: Combining the aforementioned confluence competition relationship, the flow rate of the competition constraint interval, and historical water accumulation information, calculate the node competition risk weight, and write the node competition risk weight back to the candidate edge flowing into the corresponding node to form the updated edge weight.

[0068] After obtaining the confluence competition relationship and the competition constraint flow sequence, the basic flow demand of each pipeline segment under stable transport conditions and the peak transport demand under competition superposition conditions are first determined based on the competition constraint flow sequences corresponding to each key scenario and each key time window. Specifically, in the competition constraint flow sequence, a time segment outside the key time window where the flow change tends to be stable is selected, and the representative level of the pipeline segment flow in this segment is statistically analyzed as the basic flow demand; the high-level flow in the corresponding pipeline segment flow segment within the key time window is selected, preferably the peak value or high quantile statistical value, as the peak transport demand. The flow change tends to be stable, which is judged by the flow fluctuation amplitude of multiple adjacent time steps being lower than a preset proportion; the high-level flow is determined by the high-level statistical distribution of the flow sequence within the key time window, thereby avoiding the influence of a single instantaneous outlier on the results.

[0069] After obtaining the basic flow demand and peak delivery demand, the competitive constraint interval flow rate for each pipeline segment is determined based on both. The competitive constraint interval flow rate represents the reasonable delivery range of the pipeline segment under different operating conditions, where the basic flow demand corresponds to the lower limit level under stable delivery, and the peak delivery demand corresponds to the upper limit level under the competitive superposition state. Further, the competitive constraint interval flow rate is compared with the current delivery capacity of each pipeline segment to determine the capacity constraint of each segment. When the peak delivery demand is close to or exceeds the current delivery capacity, the pipeline segment is determined to have a significant capacity constraint; when the peak delivery demand is significantly lower than the current delivery capacity, the pipeline segment is determined to have a weak capacity constraint; when the two are in the intermediate range, it is determined to be a transitional constraint state. The "close to" is determined by the difference between the peak delivery demand and the current delivery capacity being lower than a preset proportion, thus providing a clear engineering basis for capacity constraint determination.

[0070] After determining the capacity constraints of each pipe segment, based on the aforementioned confluence competition relationship, the cumulative value of the competition intensity of each candidate node under different critical scenarios is calculated. The cumulative competition intensity is the sum of the confluence competition intensities of all pairwise combinations of sub-catchments flowing towards the candidate node under the same critical scenario, used to characterize the degree of competition of the node under multi-source confluence superposition conditions. The larger the cumulative competition intensity value, the more likely the node is to experience concentrated water conveyance pressure during confluence superposition.

[0071] Furthermore, based on the capacity constraints of the pipe segments connected to each candidate node, the degree of capacity competition mismatch corresponding to each node is determined. This degree of capacity competition mismatch characterizes the mismatch between the flow demand at the node and the downstream transmission capacity. Specifically, the main control pipe segment connected to the candidate node and undertaking the primary water transmission function is read, and its peak transmission demand is compared with its current transmission capacity. When the peak transmission demand is higher than or close to the transmission capacity, it is determined to be a high mismatch state; when it is significantly lower than the transmission capacity, it is determined to be a low mismatch state; and when it is in the intermediate range, it is determined to be a medium mismatch state. The main control pipe segment is preferably the one with the largest cumulative water transmission volume within the critical time window. When the cumulative water transmission volumes of multiple pipe segments are close, the pipe segment with the longer full-flow duration is selected as the main control pipe segment.

[0072] Simultaneously, by combining the historical waterlogging information, the waterlogging sensitivity of each candidate node is determined. The historical waterlogging information is obtained through statistical analysis of historical waterlogging records. The waterlogging sensitivity is comprehensively determined by the frequency of waterlogging occurrences, duration of waterlogging, or extent of waterlogging impact in the node's area during historical rainfall events. A node is classified as a high-sensitivity node when its area experiences frequent waterlogging or prolonged waterlogging; a low-sensitivity node when there are few waterlogging records or short durations; and a moderately sensitive node when it falls between these two extremes. This process allows historical waterlogging information to participate in the risk assessment process quantitatively or semi-quantitatively.

[0073] After obtaining the cumulative competition intensity, capacity mismatch degree, and water accumulation sensitivity, the competition risk weight of each candidate node is calculated. The node competition risk weight is calculated as follows:

[0074]

[0075] Among them, among them, Indicates candidate nodes The competitive risk weight; Indicates candidate nodes Cumulative value of competition intensity under critical scenarios; This represents the normalized baseline value representing the cumulative value of competition intensity; Indicates candidate nodes The degree of capacity contention mismatch is an indicator; A normalized baseline value representing the degree of capacity contention mismatch; Indicates candidate nodes The water accumulation sensitivity index; A normalized baseline value representing the degree of waterlogging sensitivity; , , These represent the weight coefficients of each factor.

[0076] The normalized baseline value is determined by the statistical range of the cumulative competition intensity of all candidate nodes; the capacity competition mismatch index is quantified by mapping high, medium, and low mismatch states to ordered values; the water accumulation sensitivity index is quantified by mapping high, medium, and low sensitivity states to ordered values. Weighting coefficients are pre-calibrated using historical simulation samples or engineering experience to ensure that the relative impact of different factors in risk assessment conforms to the actual operating characteristics of the drainage system.

[0077] After obtaining the competition risk weights of each candidate node, these weights are written back to the candidate edges flowing into the corresponding nodes, thus updating the edge weights. Specifically, for each candidate edge, if the endpoint of the candidate edge is a candidate node... The node competition risk weight is then superimposed or merged into the original basic edge weight of the candidate edge, increasing the edge weight of the incoming edge corresponding to the node with high competition risk. This reduces the probability of the path being preferentially selected in subsequent pipeline topology optimization. For nodes with low competition risk, the edge weight of their corresponding candidate edges changes less, thus maintaining the selectability of the original path. Through this processing, the competition risk at the node level can be transmitted to the path selection at the edge level, thereby affecting the optimization direction of the subsequent pipeline topology.

[0078] In the critical scenario identification process, the intensity of node competition under different rainfall scenarios is statistically ranked to select scenarios with higher competition levels as the focus of analysis. This allows subsequent constraint analysis to concentrate on the operating conditions that have the greatest impact on the system. Compared to processing all scenarios uniformly, this selection mechanism can improve computational efficiency while avoiding interference from low-impact scenarios.

[0079] During the extraction of critical time windows, by identifying time segments where competition intensity continuously rises and remains high, the stages where multi-source superposition truly occurs are accurately captured. Compared to analyzing only peak moments, this continuous time segment identification reflects the continuity of the competition process, thus providing a more stable data foundation for subsequent traffic interval determination.

[0080] In the process of constructing the competitive constraint flow sequence, the flow changes in pipeline segments within key time windows are correlated with the corresponding competition intensity, forming a time-series data set reflecting the multi-source confluence and superposition process. Compared with the single-point flow value method, this serialized representation can retain time change information, making subsequent transportation demand determination more consistent with the actual operation process.

[0081] S4: Regenerate the pipeline topology based on updated edge weights, and construct a multi-objective evaluation index that includes pipeline cost index and expected annual loss index. Perform topology optimization iteration based on the multi-objective evaluation index until the multi-objective convergence condition is met, and output the optimized pipeline flow calculation results.

[0082] After obtaining the updated edge weights, the pipeline topology is regenerated based on the candidate pipeline network diagram. Specifically, taking the discharge outlet node as the confluence endpoint, a path search is performed from each access node to the discharge outlet node based on the updated candidate edge weights. The connected path with the smallest cumulative edge weight value is selected as the drainage path, and the above process is repeated for all access nodes to obtain the updated pipeline topology.

[0083] After generating the updated pipeline topology, for each pipeline topology, a corresponding optional pipe diameter combination is configured to form a candidate pipeline design scheme. The optional pipe diameter combination is matched and selected based on the peak transport demand of each pipe segment in the competitive constraint interval. When the peak transport demand is close to the upper limit, a larger pipe diameter specification is selected, and when it is significantly lower than the upper limit, a smaller pipe diameter specification is selected.

[0084] After formulating candidate pipeline design schemes, a multi-objective evaluation index is constructed. This multi-objective evaluation index includes a pipeline cost index and an expected annual loss index. The pipeline cost index is obtained by statistically analyzing the length and corresponding diameter of each pipe segment, and then summarizing this data in conjunction with material and construction costs.

[0085] The expected annual loss index is calculated based on overflow results under a rainy scenario, specifically:

[0086]

[0087] in, Indicates the expected annual loss indicator; Indicates a rainy scenario index; Indicates the candidate node index; Indicates the first The probability of a rainy weather scenario occurring; Indicates candidate nodes The standard for water loss in the area; Indicates the first Nodes under rainy scenarios Overflow flow rate or overflow intensity.

[0088] The probability of occurrence is obtained through historical rainfall statistics, the standard for water accumulation loss is obtained through historical water accumulation data statistics, and the node overflow is obtained by extracting overflow time series data from the hydraulic time series state data set according to the scenario.

[0089] After obtaining the pipeline cost index and the expected annual loss index, a multi-objective evaluation is performed on the candidate pipeline design schemes. Specifically, using the pipeline cost index and the expected annual loss index as dual optimization objectives, a non-dominance relationship assessment is conducted on each candidate pipeline design scheme, and solution set diversity is maintained based on the distribution of each scheme in the objective space. The non-dominance relationship assessment means that a scheme is considered to dominate the other scheme when it is not inferior to the other scheme in both objectives and is superior to the other scheme in at least one objective; the solution set diversity is achieved by controlling the distribution spacing of solutions in the objective space, so that schemes with different trade-offs can be retained.

[0090] After completing one evaluation, the candidate pipeline design schemes are screened and updated. Specifically, the candidate solution set composed of non-dominated solutions is retained, and a new round of candidate pipeline design schemes is generated based on the pipeline topology and pipe diameter combinations corresponding to the candidate solution set; hydraulic simulation and multi-objective evaluation index construction are repeatedly performed on the updated candidate schemes.

[0091] During the iteration process, a multi-objective convergence condition is used to determine whether to stop the iteration. The multi-objective convergence condition is determined by the degree of change of the candidate solution set in the target space in consecutive iterations. When the change in the number of non-dominated solution sets and the change in the target value in adjacent iterations are both lower than a preset threshold, it is determined that the convergence state has been reached; or the iteration can be terminated when the number of iterations reaches a preset upper limit.

[0092] When the iteration satisfies the multi-objective convergence condition, a Pareto front solution set composed of non-dominated solutions is output. The Pareto front solution set represents a set of optimal solutions that form different trade-offs between the pipeline cost index and the expected annual loss index. Further, based on the distribution relationship between the pipeline cost index and the expected annual loss index of each solution in the Pareto front solution set, the pipeline topology corresponding to the representative optimal solution is selected. The representative optimal solution is preferably one that achieves a balance between cost and loss, i.e., its corresponding point is located in the middle region of the Pareto front and the trend of change of adjacent solutions is relatively gentle. Finally, based on the pipeline topology corresponding to the representative optimal solution, the final pipeline flow calculation result is output.

[0093] In determining the flow range and capacity constraints, by distinguishing between stable delivery states and competing overlapping states, the basic flow demand and peak delivery demand are extracted separately, enabling the water conveyance characteristics of the pipeline segment to be expressed differently at different operating stages. Compared to using a single design flow rate, this range-based approach can avoid the problems of over-design or insufficient capacity.

[0094] By jointly analyzing competition intensity, capacity matching, and historical water accumulation information, risk assessment is based not only on current flow conditions but also on historical operational performance. Compared to methods relying solely on instantaneous flow or experience, this multi-dimensional fusion can more accurately identify potentially high-risk locations.

[0095] By writing back the node risk results to the candidate edges and regenerating the topology, path selection can proactively avoid high-risk areas. Simultaneously, multi-objective iterative screening is performed by combining cost and loss indicators. Compared to single-objective optimization methods, this approach achieves a balance between cost control and operational safety, thereby improving the rationality and stability of the overall pipeline network layout.

[0096] This embodiment also provides a multi-objective stormwater network flow calculation and optimization system based on SWMM, including:

[0097] The data processing module is used to acquire basic data of multi-source drainage and construct candidate pipe network diagrams; the simulation module is used to generate the initial pipe network topology and perform SWMM hydraulic simulation to obtain hydraulic time-series state data; the competition analysis module is used to construct the confluence competition relationship and determine the flow and capacity constraints of the competition constraint interval; the edge weight update module is used to calculate the node competition risk weight and update the edge weight of the candidate edges; the optimization module is used to generate the pipe network topology based on the updated edge weights and construct multi-objective evaluation indicators, perform topology optimization iteration and output the pipe segment flow calculation results.

[0098] This embodiment also provides a computer device applicable to the multi-objective stormwater network flow calculation and optimization method based on SWMM, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the multi-objective stormwater network flow calculation and optimization method based on SWMM as proposed in the above embodiment.

[0099] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0100] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the multi-objective stormwater network flow calculation optimization method based on SWMM as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-objective stormwater pipe network flow calculation and optimization method based on SWMM, characterized in that, include: Acquire basic data on multi-source drainage, perform spatial alignment and attribute extraction, and construct candidate pipe network diagrams; The construction of the candidate pipeline network map includes: dividing sub-catchments based on topographic elevation data; extracting the area, average slope, and impermeability of each sub-catchment; calculating the runoff demand index based on the area, slope, and impermeability of the sub-catchments; mapping each sub-catchment to its corresponding road node to form a runoff relationship; constructing a candidate pipeline network map using road intersections and outlets as nodes and road centerline segments as candidate edges; constructing basic edge weights based on the length of the candidate edges, elevation difference, historical waterlogging sensitivity, and differences in the runoff demand index of connected nodes; and performing directional processing on the candidate edges based on the node elevation relationship. An initial pipeline topology is generated based on the candidate pipeline diagram, and the initial pipeline topology is input into the SWMM model. Hydraulic simulation is performed under a rainy scenario to obtain the hydraulic time series state data of nodes and pipe segments. The hydraulic simulation under the rainy scenario includes: generating an initial pipe network topology by performing path search based on the basic edge weights of candidate edges; mapping the nodes in the initial pipe network topology to node objects in the SWMM model, mapping the candidate edges to pipe segment objects in the SWMM model, mapping the sub-catchments to sub-catchment objects in the SWMM model, configuring hydrological parameters based on the area and impermeability of the sub-catchments, constructing a rainy scenario set containing different rainfall durations and peak locations, and inputting the rainy scenario into the SWMM model; performing hydraulic simulation on the SWMM model based on the rainy scenario, obtaining the flow time series data, overflow time series data of each node, and full flow state data of each pipe segment, and forming a unified hydraulic time series state data set; Based on the hydraulic time-series state data, the confluence response segments of each sub-catchment area to the candidate nodes under different rainfall scenarios are extracted, and the confluence competition relationship is constructed according to the arrival time relationship, overlap relationship and capacity occupancy relationship between the confluence response segments; Based on the aforementioned confluence competition relationship, key scenarios and key time windows are identified, the flow rate of each pipe segment in the competition constraint interval is determined, and the capacity constraint is determined based on the flow rate in the competition constraint interval. Combining the aforementioned confluence competition relationship, the flow rate within the competition constraint interval, and historical water accumulation information, the node competition risk weight is calculated, and the node competition risk weight is written back to the candidate edges flowing into the corresponding nodes to form updated edge weights. Determining the competitive constraint interval flow includes: based on the competitive constraint flow sequence corresponding to each key scenario and each key time window, determining the basic flow demand of each pipeline segment under stable transmission state and the peak transmission demand under competitive superposition state; determining the competitive constraint interval flow of each pipeline segment according to the basic flow demand and the peak transmission demand; and determining the capacity constraint of each pipeline segment according to the relationship between the competitive constraint interval flow and the current transmission capacity of each pipeline segment. The computational node competition risk weights include: based on the confluence competition relationship, calculating the cumulative competition intensity of each node under different key scenarios; based on the capacity constraints of the pipe segments connected to each node, determining the capacity competition mismatch degree corresponding to each node; combining the historical water accumulation information, determining the water accumulation sensitivity of each node; calculating the competition risk weight of each node based on the cumulative competition intensity, capacity competition mismatch degree, and water accumulation sensitivity; and mapping the node competition risk weights to the candidate edges flowing into the corresponding nodes to increase the edge weights of the candidate edges corresponding to high competition risk areas. The pipeline topology is regenerated based on the updated edge weights, and a multi-objective evaluation index including pipeline cost index and expected annual loss index is constructed. The topology optimization iteration is performed based on the multi-objective evaluation index until the multi-objective convergence condition is met, and the optimized pipeline flow calculation results are output.

2. The multi-objective stormwater network flow calculation and optimization method based on SWMM as described in claim 1, characterized in that: The multi-source drainage basic data includes road centerline data, terrain elevation data, plot boundary data, land use data, and historical water accumulation data.

3. The multi-objective stormwater network flow calculation and optimization method based on SWMM as described in claim 2, characterized in that: The construction of the confluence competition relationship includes: based on the hydraulic time series state data set, extracting the flow response process of each sub-catchment to each candidate node under different rainfall scenarios according to a unified time step; for the flow response process of each sub-catchment at each candidate node, identifying the flow rise time, peak time and fall end time, and forming the corresponding confluence response segment. For the confluence response segments of different sub-catchments flowing toward the same candidate node, the arrival time difference, overlap duration, and composite flow occupancy during the overlap period are statistically analyzed between the segments. Based on the arrival time difference, overlap duration, and composite flow occupancy, the confluence competition intensity of different sub-catchments at the corresponding candidate node is determined. The confluence competition intensity at each candidate node under each rainfall scenario is summarized to form a confluence competition relationship that reflects the competition relationship between sub-catchments and the degree of competition between candidate nodes.

4. The multi-objective stormwater pipe network flow calculation and optimization method based on SWMM as described in claim 3, characterized in that: The identification of key scenarios and key time windows includes: based on the confluence competition relationship, calculating the cumulative competition intensity of each candidate node under different rainfall scenarios, and determining the rainfall scenario whose cumulative competition intensity meets the preset conditions as the key scenario; for the key scenario, extracting the time segment where the competition intensity at each candidate node continuously increases and remains above the preset threshold as the key time window; The sub-catchment areas and corresponding candidate nodes participating in the competition within each key time window are associated to form a competition unit under the key scenario; based on the pipe segment flow time sequence corresponding to each competition unit, the pipe segment flow segment corresponding to the key time window is extracted; based on the pipe segment flow segment and the corresponding competition intensity, a competition constraint flow sequence characterizing the multi-source confluence competition process is formed.

5. The multi-objective stormwater pipe network flow calculation and optimization method based on SWMM as described in claim 4, characterized in that: The topology optimization iteration based on the multi-objective evaluation index includes: generating candidate pipeline topologies based on the updated edge weights, and forming candidate pipeline design schemes by combining optional pipe diameter combinations; Using the pipeline cost index and the expected annual loss index as dual optimization objectives, the merits of each candidate pipeline design scheme are determined, and candidate schemes with different trade-off characteristics are retained based on their relative positions in the pipeline cost index and the expected annual loss index. Based on the merits determination results, candidate schemes that are not simultaneously degraded by other candidate schemes are retained, and a new round of candidate schemes is generated based on the retained schemes. Hydraulic simulation and multi-objective evaluation index construction are repeatedly performed on the updated candidate schemes. When the iteration meets the preset multi-objective convergence condition, the Pareto front solution set composed of the non-dominated solution set formed by the merits determination results is output. Based on the distribution relationship of each solution in the Pareto front solution set between the pipeline cost index and the expected annual loss index, the pipeline topology corresponding to the representative optimal solution is selected, and the corresponding pipe segment flow calculation results are output.

6. A multi-objective stormwater network flow calculation and optimization system based on SWMM, based on the multi-objective stormwater network flow calculation and optimization method based on any one of claims 1 to 5, characterized in that: The data processing module is used to acquire basic data of multi-source drainage and construct candidate pipe network diagrams; the simulation module is used to generate the initial pipe network topology and perform SWMM hydraulic simulation to obtain hydraulic time-series state data; the competition analysis module is used to construct the confluence competition relationship and determine the flow and capacity constraints of the competition constraint interval. The edge weight update module is used to calculate the node competition risk weight and update the edge weight of candidate edges; the optimization module is used to generate the pipeline topology based on the updated edge weights and construct multi-objective evaluation indicators, perform topology optimization iterations and output the pipeline flow calculation results.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the multi-objective stormwater pipe network flow calculation and optimization method based on SWMM as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the multi-objective stormwater pipe network flow calculation and optimization method based on SWMM as described in any one of claims 1 to 5.

Images