Port rainfall runoff simulation and waterlogging forecasting method and system based on SWMM

Abstract

The invention discloses a port rainfall runoff simulation and waterlogging forecasting method and system based on an SWMM, so as to help to improve the drainage capacity of a port area and reduce flood disasters, and the method comprises the following steps: 1, dividing simulation area sub catchment areas by using a Thiessen polygon method in ArcGIS, adjusting according to actual conditions, and exporting as an inp format; 2, setting basic parameters of the SWMM, such as rainfall recurrence period, sub catchment area parameters and the like; the SWMM model is operated, and nodes and pipe channels which are prone to overload are recognized; 3, the rainfall recurrence period is adjusted, and the maximum bearable rainfall intensity of the area is determined; and 4, LID facilities are selected for layout, an optimization algorithm is used for solving to obtain an optimal solution, and the waterlogging problem is relieved or solved. According to the method, the dynamic rainfall process is simulated based on the SWMM model, the waterlogging risks in different rainfall recurrence periods are evaluated, the waterlogging is relieved by optimally arranging the LID facilities, and support is provided for optimal management of a port drainage system.

Description

Technical Field

[0001] The invention relates to a port rainfall runoff simulation and waterlogging forecasting method and system based on SWMM, belonging to the technical field of hydrology, hydraulics and geographic information system. Background Art

[0002] With the acceleration of global climate change and urbanization, extreme weather events are frequent. Several record-breaking rainfall events in the past decade have led to unprecedented surface runoff, posing a serious threat to the safe operation of ports and other infrastructure. Runoff management is crucial to ports, but it is becoming more complex and challenging. Most traditional port drainage system designs do not take into account the collection and reuse of rainwater, so they are often unable to effectively deal with runoff peaks caused by heavy rains, leading to waterlogging and flooding in the port area. Therefore, it is particularly important to develop a system that can accurately simulate the port rainfall runoff process and optimize the drainage system management based on the simulation results.

[0003] LID (Low impact development) facilities are dispersed, small-scale facilities that purify and store rainwater at the source. Such facilities can greatly alleviate the hydrological conditions in the development area due to development, and aim to reduce the impact of land development on hydrology and the environment. LID facilities can reduce peak flow and flow at the source while reducing rainwater surface source pollution, which can effectively alleviate problems such as waterlogging and flooding in the port area.

[0004] SWMM (Storm Water Management Model) is an open source software developed by the United States Environmental Protection Agency. It is one of the most widely used and popular numerical models for managing the quantity and quality of urban runoff. The SWMM model can simulate single events or long-term precipitation runoff, and can be applied to catchments containing storm drains, combined sewer systems, sanitary sewer systems, and other urban drainage networks and features. It uses subcatchments as the basic unit of hydrological simulation, the hydraulic part is simulated by pipes and canals, and the water quality part and the hydrological part are simulated independently of each other.

[0005] There have been many application studies on SWMM in the field of simulation and analysis of LID facility deployment effects at home and abroad. SWMM is considered to be the most accurate method for modeling stormwater quality, quantity and LID performance in terms of accuracy, algorithm and covered spatiotemporal scale.

[0006] Previous studies have mainly focused on urban environments, and there are relatively few studies on the characteristics of rainfall runoff in port areas. There is a lack of in-depth analysis of rainfall runoff and its pollutant migration process, which limits the effectiveness of port water quality management. In addition, many studies lack a systematic evaluation model when proposing LID facilities, and fail to effectively compare different options, resulting in insufficient decision-making basis. Summary of the invention

[0007] In order to solve the above problems, the present invention discloses a port rainfall runoff simulation and waterlogging forecasting method and system based on SWMM, which is used for the efficient and accurate forecast of rainwater pipe network node overflow for waterlogging disasters in port areas. The specific technical scheme is as follows:

[0008] A method for simulating port rainfall runoff and forecasting waterlogging based on SWMM includes the following steps:

[0009] Step 1: Select the simulation area, determine the underlying surface type of the area, and obtain the layout data of the rainwater pipe network in the simulation area;

[0010] Step 2: Use the Thiessen polygon method in ArcGIS to divide the simulation area, obtain the sub-catchment area of ​​the simulation area, and adjust it according to the underlying surface type;

[0011] Step 3, set the basic parameters of the SWMM model, including the generation of rainfall return period and corresponding rainfall type, the setting of basic parameters of sub-catchment area, and select dynamic wave for hydraulic simulation;

[0012] Step 4, setting rainwater node parameters and rainwater pipe parameters;

[0013] Step 5, setting rainwater outlet parameters;

[0014] Step 6, optimize and adjust SWMM model parameters;

[0015] Step 7, run the SWMM model and check the overload nodes and pipes in the SWMM model simulation results;

[0016] Step 8: Change the rainfall return period, repeat step 7, compare the overload nodes and pipes with different rainfall return periods, and determine the maximum tolerable rainfall intensity in the simulation area;

[0017] Step 9, select LID facilities to be deployed in the deployable area, set LID facility parameters, and use the fast elite non-dominated sorting genetic algorithm to solve the multi-objective optimization SWMM model;

[0018] Step 10, select the optimal layout of LID facilities and repeat steps 7 to 9 until the simulation is terminated when the SWMM model output results show that waterlogging in the stormwater outlets and areas surrounding the flood-prone areas has been alleviated or eliminated.

[0019] Furthermore, in step 1, based on local hydrological conditions and topographic analysis, the underlying surface type of the simulation area is determined through on-site investigation, field experiments and local data collection, and the CAD rainwater pipe network layout data of the simulation area is obtained.

[0020] Furthermore, in step 2, the underlying surface type data of the simulation area and the CAD rainwater pipe network layout data are imported into ArcGIS, and the simulation area is divided into sub-catchment areas using the Thiessen polygon method. The node is selected as the center point, the triangulation network is connected, and the intersection points of the perpendicular bisectors of the three sides of each triangle are connected to obtain the Thiessen polygon, completing the preliminary division of the sub-catchment area; local adjustments are made according to the actual conditions of the underlying surface type, pipe network direction, buildings and street distribution, and the output is an inp file.

[0021] Furthermore, in step 3, the inp file outputted in step 2 is first imported into the SWMM model, and then the basic parameters of the SWMM model are set, specifically:

[0022] Step 3.1, set different rainfall return periods and use the Chicago Rain Type Generator to generate the corresponding rainfall type. The rainstorm intensity formula for the study area is as follows, where q is the design rainstorm intensity, L / (s·hm 2 );t is the rainfall duration, min;P is the return period, year;

[0023]

[0024] Step 3.2, set the subcatchment input parameters according to the underlying surface type obtained in step 1 and the subcatchment division result in step 2. The input parameters include the catchment area, slope, width, underlying surface type and percentage of impervious ground;

[0025] Step 3.3, select the dynamic wave method in the SWMM model for hydraulic simulation and solve it using the Saint-Venant equation, where Q is the flow rate, m 3 / s; A is the cross-sectional area of ​​water flow, m 2 ;q L is the inflow per unit length, m 3 / s; v is the flow velocity, m / s; h is the static pressure head, m; t is the time, s; x is the distance, m; g is the acceleration due to gravity, 9.8m / s 2 ; S0 is the slope of the pipe bottom; S f For friction resistance, slope descent;

[0026]

[0027] Furthermore, the step 4 of setting the rainwater node parameters and rainwater pipe parameters specifically includes:

[0028] Step 4.1, set rainwater node parameters: Rainwater node parameters include invert elevation, maximum depth and initial depth;

[0029] Step 4.2, set the rainwater pipe parameters: The rainwater pipe parameters include maximum depth, length, roughness coefficient, inlet offset and outlet offset.

[0030] Furthermore, the rainwater outlet parameters in step 5 include inflow and inner bottom elevation.

[0031] Furthermore, the step 6, optimizing and adjusting the SWMM model parameters, is specifically as follows: sensitivity analysis is performed on the SWMM model parameters, and the runoff coefficient method is combined with measured data to complete the calibration of the SWMM model hydrological and hydraulic parameters:

[0032] Step 6.1, based on the modified Morse screening method, the Chicago rain type generator is used to generate rainfall data with a return period of 3 years to conduct sensitivity analysis on the hydrological and water quality parameters of the model;

[0033] Step 6.2, based on the results of sensitivity analysis, the runoff coefficient method is used to calibrate the hydrological and hydraulic parameters of the model. If the error of the calibration result is less than 5%, it proves that the calibration result is successful and credible.

[0034] Furthermore, the results of the node overload in step 7 include the overload time, the maximum height above the top and the minimum depth below the edge, and the results of the channel overload include the full pipe hours at both ends, the full flow hours upstream and downstream and the capacity limit hours.

[0035] Furthermore, the step 9 of finding the optimal LID facility layout is specifically as follows:

[0036] Step 9.1, determine the maximum deployable area and constrain the deployment of LID facilities;

[0037] LID facility area constraints, where A i，max is the upper limit of the area that can be set for the i-th type of LID facility, m 3 .

[0038] 0≤A i ≤A i，max

[0039] Subcatchment area constraints, where is the sum of the areas of the four types of LID facilities in the jth subcatchment; A j is the area of ​​the jth subcatchment,

[0040]

[0041] Step 9.2, setting the objective function of LID layout, and constructing a multi-objective optimization model for LID layout with optimal runoff control, optimal pollutant control and minimum cost as the objective function;

[0042] Runoff control function, where Q 原，i is the rainwater runoff before the transformation of the i-th LID plot, m 3 ;Q LID，i is the rainwater runoff after the transformation of the i-th LID plot, m 3 ,

[0043] f1=∑[(Q 原，i -Q LID，i )]

[0044] Cost control function, where i is the i-th LID facility; A i is the area of ​​the i-th LID facility, m 2 ; C i is the unit price of the i-th LID facility, yuan / m 2 ; W i is the annual maintenance cost of the i-th LID facility, yuan / m 2 / Year,

[0045] f2=-∑(p×(1+e i )×A i ×C i +W i ×A i );

[0046] Pollution control function, where N 原，i is the amount of pollutants before the transformation of the i-th LID plot, kg; N LID，i is the amount of pollutants after the transformation of the i-th LID plot, kg,

[0047] f3=Σ[(N 原，i -N LID，i )]

[0048] Step 9.3, using the fast elite non-dominated sorting genetic algorithm NSGA-II to solve the multi-objective optimization model. The fast elite non-dominated sorting genetic algorithm NSGA-II generates an initial population, non-dominated sorting, crowding calculation, selection, crossover and mutation operations, and iteratively optimizes to find the Pareto front solution in the multi-objective optimization problem. The specific process is as follows:

[0049] Initial population generation: randomly generate a population, and the generated initial population P contains n individuals;

[0050] Fitness evaluation: For each individual p in the population i, calculate the objective function value, where f1, f2, and f3 are the runoff control function, cost control function, and pollution control function in step 9.2 respectively;

[0051]

[0052] Non-dominated sorting: For each individual p in the population i , calculate its dominance relationship, individual p i Dominate p j If and only if f k (p i )≤f k (p i ) holds for all k∈{1,2,3}, and there exists m∈{1,2,3} such that f m (p i ) <f m (p j ) holds true, where f1, f2, and f3 are the runoff control function, cost control function, and pollution control function in step 9.2, respectively, and k and m refer to the subscripts of functions f1, f2, and f3;

[0053] Crowding calculation: The crowding distance is calculated for each individual in the non-dominated layer to represent the diversity of individuals, where and are the maximum and minimum values ​​of the kth target, respectively.

[0054]

[0055] Selection: Combining non-dominated sorting and crowding, select individuals for crossover and mutation, give priority to individuals in the non-dominated frontier, and if the number is insufficient, select individuals with higher crowding;

[0056] Crossover and mutation: Use the simulated binary crossover (SBX) operator to generate new individuals, introduce diversity through the mutation operator, and finally obtain the offspring p', where p1 and p2 are the two selected individuals;

[0057]

[0058] Elite strategy parent-child generation merging: After the initial population is merged with the child population, the fast non-dominated sorting and crowding calculation are completed again. The selection starts from the first layer according to the crowding comparison until N individuals are selected to serve as the new parent population.

[0059] The Pareto front solution is obtained by iteration: the new parent population is selected, crossed, and mutated to generate the child population. After merging the child and parent populations, the next cycle is entered, and the steps of fitness evaluation, non-dominated sorting, crowding calculation, selection, and elite strategy parent and child merging are repeated. After multiple iterations, a set of solutions that achieve the best trade-off between multiple objectives is obtained.

[0060] The system for realizing the above-mentioned port rainfall runoff simulation and waterlogging forecasting method based on SWMM includes:

[0061] Data collection and processing module: responsible for collecting rainfall data, terrain data, land use data, drainage system data, etc. in the port area, and preprocessing and integrating the data to form basic data for SWMM model simulation;

[0062] SWMM model simulation module: Use the Thiessen polygon method to divide the simulation area into sub-catchments with reference to the actual underlying surface, and form the basic model inp file according to the input file format requirements of the SWMM software; set the basic parameters of the SWMM model and optimize and adjust them;

[0063] Result analysis module: simulate the SWMM model to obtain the conditions of overloaded nodes and overloaded pipes, and adjust the rainfall return period to determine the maximum tolerable rainfall intensity of the area;

[0064] Drainage system optimization management module: carry out LID layout, set objective functions and constraints, and use multi-objective algorithms to optimize the layout of port area rainwater utilization facilities.

[0065] The beneficial effects of the present invention are:

[0066] 1. The present invention uses the SWMM model to construct a method for the stormwater runoff system in the port area. Through dynamic simulation by SWMM, the water accumulation situation in the surrounding areas of the flood-prone points can be accurately obtained, providing guidance for improving stormwater pipes and LID facility renovation.

[0067] 2. The present invention takes into account the dynamic rainfall process and the linkage braking with the port area rainwater pipes in the SWMM model to balance the water volume and momentum, simulates and evaluates the rainwater receiving capacity of the port area under different rainfall return periods, realizes the simulation of overloaded pipes and nodes in the port area, and alleviates the waterlogging in the port area by laying out LID facilities, which more accurately reflects the drainage conditions before and after the deployment of LID facilities in the port area.

[0068] 3. The present invention reflects the waterlogging situation in the port area by evaluating the dynamic rainfall and flooding process. It can simulate the runoff changes after the deployment of LID facilities in the port area, provide forecasts for waterlogging in areas prone to waterlogging, and reduce the losses caused by flood disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1is a flow chart of the present invention;

[0070] Figure 2a It is the sub-catchment parameter content in the inp file created in the implementation of the present invention;

[0071] Figure 2b It is the pipe parameter content in the inp file created in the implementation of the present invention;

[0072] Figure 2c It is the node parameter content in the inp file created in the implementation of the present invention;

[0073] Figure 2d It is the discharge port parameter content in the inp file created in the implementation of the present invention;

[0074] Figure 3a It is the result of overloaded nodes in the inp file created in the implementation of the present invention;

[0075] Figure 3b It is the result of overloaded pipes and canals in the inp file created in the implementation of the present invention;

[0076] Figure 4 The model file created in the embodiment of the present invention is displayed in the SWMM model. DETAILED DESCRIPTION

[0077] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0078] The present invention performs real-time simulation based on the SWMM model, which can predict the overflow volume and overflow position of the future pipe network nodes, thereby simulating the rainfall runoff process in the port area. Optimization management based on these results can effectively improve the drainage capacity of the port area and alleviate problems such as waterlogging and flooding in the port area. The multi-objective optimization model is adopted, combined with the three major goals of runoff control, pollutant control and cost, to provide a scientific basis for the layout of LID facilities. This method is innovative in comprehensive decision-making and can provide specific solutions for practical applications.

[0079] Combined with Figure 1 It can be seen that the present invention specifically comprises the following steps:

[0080] Taking the Phase 3 yard terminal of a coal port in the north as an example, the area is about 108 hectares, with a high degree of development, moderate scope, and detailed pipe network layout and port land type classification data, and there are significant differences in the functions of different zones. The area is located in the warm temperate coastal semi-humid continental monsoon climate type area, with an average annual precipitation of 616.8 mm and a maximum annual precipitation of 931.7 mm. The precipitation is concentrated in July and August, accounting for 60% of the annual precipitation.

[0081] Step 1: Select the simulation area, determine the regional underlying surface type and obtain the regional CAD rainwater pipe network layout data based on local hydrological conditions and topographic analysis through on-site investigation, field tests and background data collection.

[0082] Step 2: Import the data of the simulation area and the CAD rainwater pipe network file into ArcGIS, divide the simulation area into sub-catchments using the Thiessen polygon method, and make manual adjustments based on the actual underlying surface, and output them as inp files. The selected area has a total of 198 nodes, 198 sub-catchments, 3 rainwater outlets, and 199 pipes and canals, of which the office area is all rainwater pipes, and the yard and work area are rainwater pipes and ditches.

[0083] Step 3, import the inp file into the SWMM model and set the basic parameters of the SWMM model.

[0084] Step 3.1, determine the rainfall recurrence period, and use the Chicago rain pattern generator to generate the corresponding rain pattern. In this embodiment, the rainfall duration is 2 hours, the time interval is 1 minute, the rain peak coefficient is selected as 0.4, and the recurrence periods are 1, 3, 5, 10, 30, and 50 years respectively.

[0085] Step 3.2: Set the basic parameters of the subcatchment based on the data obtained in step 1. The input parameters include the catchment area, slope, width, underlying surface type, and percentage of impervious ground, etc. Figure 2a shown.

[0086] In step 3.3, select the dynamic wave equation in SWMM for hydraulic simulation.

[0087] Step 4: Set the rainwater node parameters and rainwater pipe parameters.

[0088] Step 4.1, Rainwater node parameter setting: Input parameters include invert elevation, maximum depth and initial depth, such as Figure 2b shown.

[0089] Step 4.2, rainwater pipe parameter setting: input parameters include maximum depth, length, roughness coefficient, inlet offset and outlet offset, etc. Figure 2c shown.

[0090] Step 5, rainwater outlet parameter setting: input parameters include inflow, inner bottom elevation, etc. Figure 2d shown.

[0091] Step 6: Optimize and adjust SWMM parameters, conduct sensitivity analysis on model parameters, and use the runoff coefficient method combined with measured data to calibrate the SWMM model's hydrological and hydraulic parameters.

[0092] Step 6.1, based on the modified Morse screening method, the Chicago rain type generator is used to generate rainfall data with a return period of 3 years to conduct sensitivity analysis on the hydrological and water quality parameters of the model.

[0093] In step 6.2, based on the results of sensitivity analysis, the runoff coefficient method is used to calibrate the hydrological and hydraulic parameters of the model. The error of the calibration result is less than 5%, proving that the calibration result is successful and credible.

[0094] Step 7: Run SWMM simulation and check the overloaded nodes and overloaded pipes in the simulation results. The results of overloaded nodes include overload time, maximum height above the top, and minimum depth below the edge. The results of pipe overload include full pipe hours at both ends, full flow hours upstream and downstream, and capacity limit hours. The results of overloaded nodes and overloaded pipes are as follows: Figure 3a , Figure 3b shown.

[0095] Step 8: Change the rainfall return period and repeat step 7 to compare the overloaded nodes and overloaded pipes at different rainfall return periods to determine the maximum tolerable rainfall intensity of the area.

[0096] Step 9: Select LID facilities and deploy them in the deployable area to find the optimal LID facility deployment.

[0097] Step 9.1, determine the maximum deployable area.

[0098] Step 9.2, set the objective function of LID layout, and build a multi-objective optimization model for LID layout with optimal runoff control, optimal pollutant control and minimum cost as the objective functions.

[0099] Step 9.3, use the fast elite non-dominated sorting genetic algorithm (NSGA-II) to solve the multi-objective optimization model. Among them, NSGA-II can iteratively optimize to find the Pareto front solution in the multi-objective optimization problem through generating the initial population, non-dominated sorting, crowding calculation, selection, crossover and mutation operations.

[0100] Step 10, select the optimal layout of LID facilities, repeat steps 7 to 9, and end the simulation when the model output results show that waterlogging in the stormwater outlets and the surrounding areas of the flood-prone areas has been alleviated or eliminated.

[0101] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0102] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A port rainfall runoff simulation and waterlogging forecasting method based on SWMM, characterized in that: The following steps are involved: Step 1: Select the simulation area, determine the underlying surface type of the area, and obtain the layout data of the rainwater pipe network in the simulation area; Step 2: Use the Thiessen polygon method in ArcGIS to divide the simulation area, obtain the sub-catchment area of ​​the simulation area, and adjust it according to the underlying surface type; Step 3, set the basic parameters of the SWMM model, including the generation of rainfall return period and corresponding rainfall type, the setting of basic parameters of sub-catchment area, and select dynamic wave for hydraulic simulation; Step 4, setting rainwater node parameters and rainwater pipe parameters; Step 5, setting rainwater outlet parameters; Step 6, optimize and adjust SWMM model parameters; Step 7, run the SWMM model and check the overload nodes and pipes in the SWMM model simulation results; Step 8: Change the rainfall return period, repeat step 7, compare the overload nodes and pipes with different rainfall return periods, and determine the maximum tolerable rainfall intensity in the simulation area; Step 9, select LID facilities to be deployed in the deployable area, set LID facility parameters, and use the fast elite non-dominated sorting genetic algorithm to solve the multi-objective optimization SWMM model; Step 10, select the optimal layout of LID facilities and repeat steps 7 to 9 until the simulation is terminated when the SWMM model output results show that waterlogging in the stormwater outlets and areas surrounding the flood-prone areas has been alleviated or eliminated.

2. The method for simulating port rainfall runoff and forecasting waterlogging based on SWMM according to claim 1, characterized in that: In step 1, based on the local hydrological conditions and topographic analysis, the underlying surface type of the simulation area is determined through on-site investigation, field tests and local data collection, and the CAD rainwater pipe network layout data of the simulation area is obtained.

3. The method for simulating port rainfall runoff and predicting waterlogging based on SWMM according to claim 1, characterized in that: In the step 2, the underlying surface type data of the simulation area and the CAD rainwater pipe network layout data are imported into ArcGIS, and the sub-catchment area of ​​the simulation area is divided using the Thiessen polygon method. The node is selected as the center point, the triangulation network is connected, and the intersection points of the perpendicular bisectors of the three sides of each triangle are connected to obtain the Thiessen polygon, completing the preliminary division of the sub-catchment area; local adjustments are made according to the actual conditions of the underlying surface type, pipe network direction, buildings and street distribution, and the output is an inp file.

4. The port rainfall runoff simulation and waterlogging forecasting method based on SWMM according to claim 3 is characterized in that: In step 3, the inp file output in step 2 is first imported into the SWMM model, and then the basic parameters of the SWMM model are set, specifically: Step 3.1, set different rainfall return periods and use the Chicago Rain Type Generator to generate the corresponding rainfall type. The rainstorm intensity formula for the study area is as follows, where q is the design rainstorm intensity, L / (s·hm 2 );t is the rainfall duration, min;P is the return period, year; Step 3.2, set the subcatchment input parameters according to the underlying surface type obtained in step 1 and the subcatchment division result in step 2. The input parameters include the catchment area, slope, width, underlying surface type and percentage of impervious ground; Step 3.3, select the dynamic wave method in the SWMM model for hydraulic simulation and solve it using the Saint-Venant equation, where Q is the flow rate, m 3 / s; A is the cross-sectional area of ​​water flow, m 2 ;q L is the inflow per unit length, m 3 / s; v is the flow velocity, m / s; h is the static pressure head, m; t is the time, s; x is the distance, m; g is the acceleration due to gravity, 9.8m / s 2 ; S0 is the slope of the pipe bottom; S f For friction resistance, the slope is lowered; 5. The method for simulating port rainfall runoff and forecasting waterlogging based on SWMM according to claim 1, characterized in that: The step 4, setting the rainwater node parameters and rainwater pipe parameters specifically includes: Step 4.1, set rainwater node parameters: Rainwater node parameters include invert elevation, maximum depth and initial depth; Step 4.2, set the rainwater pipe parameters: The rainwater pipe parameters include maximum depth, length, roughness coefficient, inlet offset and outlet offset.

6. The method for simulating port rainfall runoff and forecasting waterlogging based on SWMM according to claim 1, characterized in that: The rainwater outlet parameters in step 5 include inflow and inner bottom elevation.

7. The method for simulating port rainfall runoff and forecasting waterlogging based on SWMM according to claim 1, characterized in that: The step 6, optimizing and adjusting the SWMM model parameters, is specifically: performing sensitivity analysis on the SWMM model parameters, and using the runoff coefficient method in combination with measured data to complete the calibration of the SWMM model hydrological and hydraulic parameters: Step 6.1, based on the modified Morse screening method, the Chicago rain type generator is used to generate rainfall data with a return period of 3 years to conduct sensitivity analysis on the hydrological and water quality parameters of the model; Step 6.2, based on the results of sensitivity analysis, the runoff coefficient method is used to calibrate the hydrological and hydraulic parameters of the model. If the error of the calibration result is less than 5%, it proves that the calibration result is successful and credible.

8. The method for simulating port rainfall runoff and forecasting waterlogging based on SWMM according to claim 1, characterized in that: The results of node overload in step 7 include overload time, maximum height above the top and minimum depth below the edge, and the results of pipeline overload include full pipe hours at both ends, full flow hours upstream and downstream and capacity limit hours.

9. The method for simulating port rainfall runoff and forecasting waterlogging based on SWMM according to claim 1, characterized in that: The step 9 of finding the optimal LID facility layout is specifically as follows: Step 9.1, determine the maximum deployable area and constrain the deployment of LID facilities; LID facility area constraints, where A i,max is the upper limit of the area that can be set for the i-th type of LID facility, m 3 . 0≤A i ≤A i,max Subcatchment area constraints, where is the sum of the areas of the four types of LID facilities in the jth subcatchment; A j is the area of ​​the jth subcatchment, Step 9.2, setting the objective function of LID layout, and constructing a multi-objective optimization model for LID layout with optimal runoff control, optimal pollutant control and minimum cost as the objective function; Runoff control function, where Q 原,i is the rainwater runoff before the transformation of the i-th LID plot, m 3 ;Q LID,i is the rainwater runoff after the transformation of the i-th LID plot, m 3 , f1=∑[(Q 原,i -Q LID,i )]; Cost control function, where i is the i-th LID facility; A i is the area of ​​the i-th LID facility, m 2 ; C i is the unit price of the i-th LID facility, yuan / m 2 ; W i is the annual maintenance cost of the i-th LID facility, yuan / m 2 / Year, f2=-∑(p×(1+e i )×A i ×C i +W i ×A i ); Pollution control function, where N 原,i is the amount of pollutants before the transformation of the i-th LID plot, kg; N LID,i is the amount of pollutants after the transformation of the i-th LID plot, kg, <h2 style=";text-align:left;direction:ltr">f3=∑[(N<h2 style=";text-align:left;direction:ltr"> 原,i <h2 style=";text-align:left;direction:ltr"> -N<h2 style=";text-align:left;direction:ltr"> LID,i <h2 style=";text-align:left;direction:ltr"> )] Step 9.3, using the fast elite non-dominated sorting genetic algorithm NSGA-II to solve the multi-objective optimization model. The fast elite non-dominated sorting genetic algorithm NSGA-II generates an initial population, non-dominated sorting, crowding calculation, selection, crossover and mutation operations, and iteratively optimizes to find the Pareto front solution in the multi-objective optimization problem. The specific process is as follows: Initial population generation: randomly generate a population, and the generated initial population P contains n individuals; Fitness evaluation: For each individual p in the population i , calculate the objective function value, where f1, f2, and f3 are the runoff control function, cost control function, and pollution control function in step 9.2 respectively; Non-dominated sorting: For each individual p in the population i , calculate its dominance relationship, individual p i Dominate p j If and only if f k (p i )≤f k (p j ) holds for all k∈{1,2,3}, and there exists m∈{1,2,3} such that f m (p i ) <f m (p j ) holds, where f1, f2, f3 are the runoff control function, cost control function and pollution control function in step 9.2 respectively. Both k and m refer to the subscripts of functions f1, f2, and f3; Crowding calculation: The crowding distance is calculated for each individual in the non-dominated layer to represent the diversity of individuals, where and are the maximum and minimum values ​​of the kth target, respectively. Selection: Combining non-dominated sorting and crowding, select individuals for crossover and mutation, give priority to individuals in the non-dominated frontier, and if the number is insufficient, select individuals with higher crowding; Crossover and mutation: Use the simulated binary crossover (SBX) operator to generate new individuals, introduce diversity through the mutation operator, and finally obtain the offspring p', where p1 and p2 are the two selected individuals; Elite strategy parent-child generation merging: After the initial population is merged with the child population, the fast non-dominated sorting and crowding calculation are completed again. The selection starts from the first layer according to the crowding comparison until N individuals are selected to serve as the new parent population. The Pareto front solution is obtained by iteration: the new parent population is selected, crossed, and mutated to generate the child population. After merging the child and parent populations, the next cycle is entered, and the steps of fitness evaluation, non-dominated sorting, crowding calculation, selection, and elite strategy parent and child merging are repeated. After multiple iterations, a set of solutions that achieve the best trade-off between multiple objectives is obtained.

10. A system for implementing the SWMM-based port rainfall runoff simulation and waterlogging forecasting method according to any one of claims 1 to 9, characterized in that: include: Data collection and processing module: responsible for collecting rainfall data, terrain data, land use data, drainage system data, etc. in the port area, and preprocessing and integrating the data to form basic data for SWMM model simulation; SWMM model simulation module: Use the Thiessen polygon method to divide the simulation area into sub-catchments with reference to the actual underlying surface, and form the basic model inp file according to the input file format requirements of the SWMM software; set the basic parameters of the SWMM model and optimize and adjust them; Result analysis module: simulate the SWMM model to obtain the conditions of overloaded nodes and overloaded pipes, and adjust the rainfall return period to determine the maximum tolerable rainfall intensity of the area; Drainage system optimization management module: carry out LID layout, set objective functions and constraints, and use multi-objective algorithms to optimize the layout of port area rainwater utilization facilities.

Images