A sewer network potential infiltration risk area intelligent grading identification method

Abstract

The application provides a kind of sewer potential infiltration risk area intelligent grading identification method, it is related to sewer risk identification technical field, this method constructs surface water-groundwater coupling model and sewer model respectively for simulation calculation, spatial relationship mapping is established between the triangular grid element of surface water-groundwater coupling model and the pipe section of sewer model, the hourly water level difference time series data is calculated using the groundwater level time series data and the corresponding pipe section liquid level time series data obtained by simulation calculation, and a risk index is constructed accordingly;Then an uncertainty scenario sample set is constructed and input into the above two models respectively to obtain the corresponding simulation results, the risk index corresponding to each pipe section is calculated, and the grading result of the potential infiltration risk area is determined according to the risk index. The present application combines simulation calculation and intelligent diagnostic analysis, and realizes efficient identification of potential infiltration risk area of sewer without large-scale sensor deployment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sewer network risk identification, and particularly relates to an intelligent hierarchical identification method for a potential infiltration risk area of a sewer network. BACKGROUND

[0002] A sewer network mainly includes a rainwater network, a sewage network and its auxiliary facilities. The rainwater pipeline collects and transports rainwater, and the sewage pipeline is responsible for collecting and transporting urban sewage. As an important part of urban infrastructure, the operation status of the sewer network is directly related to the urban environmental quality. However, due to the initial design problems, long-term use and aging and the influence of external environmental factors, the current underground water infiltration problem caused by the damage of the sewer network has become a bottleneck restricting the normal operation of the sewer network, which seriously affects the operation efficiency of the sewer network and increases the burden of sewage treatment; especially in the areas with abundant rainwater and high groundwater level, the problem of clean water invading the sewage pipeline is prominent.

[0003] In view of the underground water infiltration problem caused by the damage of the sewer network, the intelligent diagnosis method for the potential infiltration risk area of the sewer network is designed, which not only helps to identify the risk problem area, but also provides a basis for subsequent risk early warning, operation and maintenance scheduling and preventive management measures, and has important significance for guaranteeing the sustainable development of the urban drainage system. However, the identification of the potential infiltration risk area of the sewer network still faces many challenges. The sewer network is usually buried underground, and the system structure is complex, large in scale and highly coupled with other underground infrastructure, and the spatial layout is staggered. Although the traditional detection means including CCTV endoscopic detection and liquid level sensor arrangement have feasibility in the local area, due to the high deployment cost, large demand for manpower and the limitation of the detection results on the accessibility of the pipeline, the size of the pipeline and the siltation condition, it is difficult to realize the full coverage of the regional level. At the same time, the sewer network system has significant time-varying and uncertainty, that is, even if the current state of the pipeline section is good, the risk of infiltration may occur in a short period of time, so that it is difficult to reflect the real risk level of the system by relying on one-time detection, and it is impossible to capture the possible risk in the future. In addition, the defects of the pipeline structure itself are hidden, so even if the local detection is normal, the potential infiltration risk may exist, which makes it difficult for the traditional means to realize the forward-looking and continuous risk identification. These dynamic evolution characteristics greatly increase the technical challenges of identifying high-risk areas, and also strengthen the urgency of building a system with dynamic simulation and intelligent discrimination ability.

[0004] The existing technology mainly focuses on the equipment update of the "front-end" detection means, such as the improvement of sensor performance and the improvement of video recognition technology, and the comprehensive analysis of "back-end" data and the diagnosis research based on mechanism deduction are still obviously insufficient. Especially under the realistic condition that the monitoring points are sparse and the coverage is limited, it is difficult to support the planar potential infiltration risk identification by relying only on the monitoring data, and the system and intelligence are lacking. SUMMARY

[0005] The application aims to provide an intelligent grading identification method for potential infiltration risk areas of a drainage pipe network based on a surface water-groundwater coupling model and a drainage pipe network model.

[0006] Technical scheme: An intelligent grading identification method for potential infiltration risk areas of a drainage pipe network comprises the following steps:

[0007] S1, obtain the basic data of a target area, construct and calibrate an initial surface water-groundwater coupling model and an initial drainage pipe network model respectively, and obtain a target surface water-groundwater coupling model and a target drainage pipe network model;

[0008] S2, use a GIS spatial analysis tool to establish the corresponding or adjacent relationship between each triangular grid cell of the target surface water-groundwater coupling model and each pipe section of the target drainage pipe network model based on spatial position;

[0009] S3, close the groundwater simulation module in the target drainage pipe network model, input meteorological data and typical sewage discharge time series characteristic data in the basic data to drive the target drainage pipe network model to run, and obtain the liquid level time series data of each pipe section in the target area; use the same meteorological data to drive the target surface water-groundwater coupling model to run in parallel, and obtain the groundwater level time series data of the triangular grid cells corresponding to each pipe section;

[0010] S4, use the corresponding or adjacent relationship between each triangular grid cell and each pipe section, calculate the liquid level difference data of each triangular grid cell by subtracting the liquid level of the corresponding pipe section from the groundwater level of the triangular grid cell, construct the hourly liquid level difference time series data, and construct the risk index of each pipe section according to the liquid level difference time series data;

[0011] S5, set several key control factors, and construct an uncertainty scenario sample set by perturbing these key control factors;

[0012] S6, input the uncertainty scenario sample set into the target surface water-groundwater coupling model and the target drainage pipe network model respectively for simulation, and obtain the groundwater level and pipe section liquid level time series results under each scenario; use the results for the risk index calculation in S4, obtain the risk index of each pipe section under all uncertainty scenarios, and determine the grading results of the potential infiltration risk area according to the risk index of each pipe section.

[0013] Specifically, in step S1, the basic data includes basic data of the initial surface water-groundwater coupling model and basic data of the initial drainage pipe network model; the basic data of the initial surface water-groundwater coupling model includes elevation data, a research range, a pipe network distribution range, underlying surface attribute data, meteorological data, and calibration data; and the basic data of the initial drainage pipe network model includes typical sewage discharge time sequence characteristic data of different land use types in the research area, water conservancy structure data, drainage zoning data, pipe network data, population statistics data, elevation data, underlying surface attribute data, sub-catchment division data, meteorological data, and calibration data.

[0014] Specifically, in step S1, the initial surface water-groundwater coupling model is constructed based on inteliway-SSIM software, and the steps of constructing and calibrating the initial surface water-groundwater coupling model are as follows:

[0015] According to the elevation data, the research range, and the pipe network distribution range, the target area is divided into triangular grid cells in inteliway-SSIM;

[0016] Based on the elevation data and the underlying surface attribute data, each triangular grid cell is assigned with corresponding attribute parameters;

[0017] The meteorological data is input to drive inteliway-SSIM to run, and the simulation result of the groundwater level data is obtained;

[0018] The calibration data is the historical monitoring data of the groundwater level, the simulation result of the groundwater level is compared with the historical monitoring data of the groundwater level, if the error between them exceeds a set threshold, the model parameters of inteliway-SSIM are adjusted, or the grid is re-divided and re-simulated, until the error does not exceed the set threshold.

[0019] Specifically, in step S1, the initial drainage pipe network model is constructed based on inteliway-SWMM software, and the steps of constructing and calibrating the initial drainage pipe network model are as follows:

[0020] According to the elevation data and the underlying surface attribute data, each sub-catchment obtained by division is assigned with corresponding attribute values, and the corresponding rainfall time sequence of the nearest meteorological station is extracted according to the geographical range of each sub-catchment;

[0021] The meteorological data and the typical sewage discharge time sequence characteristic data are input to drive inteliway-SWMM to run, and the simulation result of the flow, liquid level, and water quality of each pipe segment is obtained;

[0022] The calibration data are historical monitoring data of the flow, liquid level and water quality of each pipe section. The simulation results of the flow, liquid level and water quality of each pipe section are compared with the historical monitoring data of the flow, liquid level and water quality of each pipe section. If the error exceeds the set threshold, the model parameters of inteliway-SWMM are adjusted and the simulation is repeated until the error does not exceed the set threshold.

[0023] Specifically, in step S4, the risk index of each pipe section is constructed as follows:

[0024] Based on the time series data of the liquid level difference, the liquid level difference threshold indicating the potential infiltration risk is set, and the risk index of the pipe section is constructed. The risk index includes the average water level difference, the exceeding time proportion and the cumulative excess high water head.

[0025] The average water level difference calculation formula is:

[0026]

[0027]

[0028] In the formula: is the average water level difference, the unit is meter, n is the total number of time points meeting t represents the tth time point in the simulation period, T is the total number of time points in the simulation period, is the liquid level difference at the tth time point, the unit is meter, is the liquid level difference threshold indicating the potential infiltration risk, the unit is meter, is the groundwater level at the tth time point, the unit is meter, is the pipe section liquid level at the tth time point, the unit is meter;

[0029] The exceeding time proportion calculation formula is:

[0030]

[0031]

[0032] In the formula: is the exceeding time proportion, is the indicator function;

[0033] The cumulative excess high water head calculation formula is:

[0034]

[0035]

[0036] In the formula: is the cumulative excess high water head, the unit is meter, is the time step, the unit is hour.

[0037] Specifically, step S5 includes the following sub-steps:

[0038] S51, select target surface water-groundwater coupling model parameters, rainfall boundary factors and pollution discharge boundary factors as key control factors, the target surface water-groundwater coupling model parameters include permeability coefficient, porosity, recharge rate and discharge coefficient, the rainfall boundary factors include storm recurrence period, rainfall duration and intensity-duration-frequency curve parameters, the pollution discharge boundary factors include unit population pollution discharge intensity and pollution discharge daily variation law;

[0039] S52, set a corresponding disturbance mode for different key control factors;

[0040] S53, set the disturbance interval of different key control factors according to historical monitoring data or expert knowledge;

[0041] S54, apply disturbance to the key control factors to generate key control factor disturbance samples, and perform physical rationality check and parameter constraint relationship check on the samples, and obtain the initial disturbance sample set of each key control factor after eliminating unqualified samples;

[0042] S55, sample and combine the initial disturbance sample set of each key control factor to construct an uncertainty scenario sample set.

[0043] Specifically, step S52 includes:

[0044] The disturbance interval and probability distribution form of the target surface water-groundwater coupling model parameters are determined by using a distribution fitting method;

[0045] The statistical characteristics of the rainfall boundary factors are inferred according to historical meteorological data of the target region, statistical modeling is performed in combination with historical storm measurement data of the target region, and the rainfall process is empirically disturbed according to the storm classification standard;

[0046] The unit population pollution discharge intensity is disturbed by using a normal distribution or a uniform distribution, and a number of representative daily variation factor sequences are generated by randomly selecting from a typical daily variation curve sample set.

[0047] Specifically, step S55 includes:

[0048] S551, perform Monte Carlo high-density sampling on the initial disturbance sample set of the target surface water-groundwater coupling model parameters to obtain an initial parameter sample set, perform normalization preprocessing on the initial parameter samples, then use a clustering algorithm to cluster the initial parameter sample set, extract a number of samples from each category from the clustering result, perform inverse normalization on the samples, and use the inverse normalized samples to construct a target surface water-groundwater coupling model parameter disturbance sample set;

[0049] S552, Latin hypercube sampling method is used on the initial perturbation sample set of the rainfall boundary factor to construct the perturbation sample set of the rainfall boundary factor;

[0050] S553, according to the unit population pollution intensity, the daily total sewage quantity time series data is obtained, combined with several groups of representative daily variation factor sequences and corresponding daily total sewage quantity data, the daily sewage discharge time series is generated through the hourly multiplication operation to construct the initial perturbation sample set of the pollution discharge boundary factor;

[0051] S554, the obtained target surface water-groundwater coupling model parameter perturbation sample set, rainfall boundary factor perturbation sample set and pollution discharge boundary factor initial perturbation sample set are combined to obtain an uncertainty scenario sample set, and the sample expression in the uncertainty scenario sample set is:

[0052]

[0053] In the formula: is a certain sample in the uncertainty scenario sample set, is the i th sample in the target surface water-groundwater coupling model parameter perturbation sample set, is the j th sample in the rainfall boundary factor perturbation sample set, is the k th sample in the pollution discharge boundary factor perturbation sample set.

[0054] Specifically, in step S6, the step of determining the grading result of the potential infiltration risk area according to the risk index of each pipe section is as follows: the risk index is normalized, and a comprehensive risk index is set, and the calculation formula of the comprehensive risk index is:

[0055]

[0056] In the formula: is the comprehensive risk index of the i th pipe section under all uncertainty scenarios, , and is the set weight, is the normalized average water level difference of the i th pipe section under all uncertainty scenarios, is the normalized over-standard time proportion of the i th pipe section under all uncertainty scenarios, is the normalized cumulative over-high water head of the i th pipe section under all uncertainty scenarios;

[0057] According to the comprehensive risk index result, the quantile method is used to determine and divide the potential infiltration risk level of each pipe section in combination with the set threshold value based on expert experience.

[0058] Specifically, the quantile method is adopted, and the threshold value is set based on expert experience to determine and divide the potential infiltration risk level of each pipe section as follows:

[0059] Collect the comprehensive risk indicators of all pipe sections under all uncertainty scenarios to form a risk indicator dataset, divide the dataset into several quantile intervals using the quantile method, and map each quantile interval to a predetermined risk level;

[0060] Adjust the upper and lower threshold values of the quantile intervals obtained based on expert experience;

[0061] Map the comprehensive risk indicators of each pipe section to the corresponding risk level using the adjusted quantile intervals to obtain the potential infiltration risk level classification results of all pipe sections.

[0062] Advantages: Compared with the prior art, the significant effects of the present application are:

[0063] 1. The present application integrates multi-source data under the causal reasoning framework, uses a surface water-groundwater coupling model and a drainage pipe network model to construct a unified diagnosis framework, performs systematic simulation and potential infiltration risk identification on the pipe network in the target area, simulates the dynamic changes of regional groundwater through the surface water-groundwater coupling model, realizes intelligent identification of the potential infiltration risk area of groundwater and risk level determination by combining the liquid level difference and risk indicator calculation, makes up for the limitation that the traditional data measurement method cannot achieve spatial full coverage, improves the comprehensiveness and systematicness of potential infiltration risk identification, and greatly reduces the monitoring cost compared with the traditional manual sampling detection method and a large number of equipment layout monitoring means, improves the identification efficiency.

[0064] 2. The present application designs a three-dimensional risk index system with “average super-threshold water head difference (R1) - time ratio of over-standard (R2) - cumulative super-high water head (R3)” as the core, quantitatively describes the potential infiltration risk of groundwater from three dimensions of infiltration intensity, duration and accumulation effect, and can capture both short-term sudden leakage and long-term delayed penetration. Compared with single index or experience-based methods, the three-dimensional risk index system is more comprehensive and sensitive, and can effectively reduce the risk of missed judgment and misjudgment.

[0065] 3. The present application constructs an uncertainty scenario sample set by setting and disturbing key control factors, simulates the dynamic changes of groundwater state in time and space using the uncertainty scenario sample set, breaks through the traditional single working condition assumption, can accurately reflect the diversified response characteristics of potential infiltration risk under different conditions, and can improve the generalization ability and applicability of the potential infiltration risk evaluation results.

[0066] 4、The application constructs a comprehensive risk index on the basis of designing various risk indexes, and constructs a fusion grading mechanism of "quantile initial division-expert threshold correction-event alignment verification" according to the result of the comprehensive risk index, so as to consider the distribution characteristics of the scoring data and the reasonable judgment of the actual experience. By adopting the quantile method to determine the initial risk level boundary, and then combining the expert knowledge to adjust the key quantile point, a multi-level division system with interpretability and generalizability is constructed. The application breaks through the dependence of the traditional grading method on single data or experience, improves the adaptability and reliability of the grading result under different regions and working conditions, and provides a more scientific basis for the identification and grading management of the drainage pipe section risk.

[0067] 5、The application mainly relies on two types of basic monitoring data of pipe section liquid level and underground water level, has actual landing ability under the condition of lacking network flow meter layout, is suitable for inventory old pipe network, complex mixed system and newly-built drainage area. Meanwhile, the identification result of the potential infiltration risk area can be complementary to the existing CCTV detection, smoke test and leakage measurement and other technologies, and a multi-means combined diagnosis system is constructed.

[0068] 6、The identification result is quantized and graded and spatially and temporally positioned with the pipe section as a basic unit, has good mapping ability, and can be used for guiding the annual management plan formulation, performance evaluation and overflow control and other target management tasks. The application supports the closed-loop management logic of "monitoring-diagnosis-management-recheck", and provides a scientific basis and technical support for the fine management and operation and maintenance of the urban drainage pipe network. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 It is a method flowchart of the application.

[0070] Figure 2 It is a schematic diagram of the spatial mapping relationship of the ISSIM model and the ISWMM model units of the application.

[0071] Figure 3 It is a schematic diagram of the local underground water level spatial distribution of the target area of the application.

[0072] Figure 4 It is a curve graph of the underground water depth and rainfall time series in the target area of the application.

[0073] Figure 5 It is a schematic diagram of the local potential infiltration risk area grading identification result in the target area of the application. DETAILED DESCRIPTION

[0074] One preferred embodiment of the application will be further described below in combination with the drawings.

[0075] Please refer to Figure 1As shown, the present application provides an intelligent grading identification method for potential infiltration risk areas of a sewer network, comprising the following steps:

[0076] S1, obtain the basic data of the target area, use the basic data to construct an initial surface water-groundwater coupling model and an initial sewer network model, use the basic data to drive the initial surface water-groundwater coupling model and the initial sewer network model to run, obtain the initial simulation results, use the calibration data in the basic data and the initial simulation results to calibrate the initial surface water-groundwater coupling model, obtain the target surface water-groundwater coupling model, use the calibration data in the basic data and the initial simulation results to calibrate the initial sewer network model, obtain the target sewer network model;

[0077] The initial surface water-groundwater coupling model construction and calibration process is as follows:

[0078] The initial surface water-groundwater coupling model used in the present application is based on inteliway-SSIM software (hereinafter referred to as ISSIM), which is a three-dimensional surface water-groundwater coupling simulation software developed by Beijing Inteli for Environmental Technology Co., Ltd. This model can accurately depict the response of surface and groundwater hydrological processes under the combined action of different conditions and human activities by fully coupling surface hydrology and groundwater hydrology. The model includes the following functions: one-dimensional river channel hydrodynamic process simulation, two-dimensional surface water process simulation, three-dimensional groundwater process simulation, two-way interaction simulation of surface water and groundwater, river and surrounding surface and groundwater interaction simulation, open boundary interaction simulation and reservoir simulation. In the present application, the reservoir simulation function module is not used.

[0079] The basic data used to construct ISSIM includes elevation data (DEM), research scope, pipe network distribution range, underlying surface attribute data, meteorological data and calibration data; the underlying surface attribute data includes soil type, vegetation distribution, geological structure and land use type; the meteorological data includes rainfall time series data, temperature time series data, relative humidity time series data, wind speed time series data and radiation time series data; the calibration data is historical monitoring data of groundwater level.

[0080] According to the elevation data, the research scope and the pipe network distribution range, the research scope of the target area is divided into triangular grid units in ISSIM; based on the elevation data and the underlying surface attribute data, the triangular grid units are assigned with corresponding attributes; the meteorological data is input into ISSIM and drives ISSIM, and ISSIM outputs the initial simulation results, which are the simulation results of groundwater level data; the simulation results of groundwater level data are compared with the historical monitoring data of groundwater level, if the error exceeds the set threshold, the model parameters of ISSIM are adjusted, or the grid is re-divided and re-simulated until the error does not exceed the set threshold.

[0081] The initial drainage pipe network model construction and calibration process is as follows:

[0082] The initial drainage pipe network model adopted in the present application is based on inteliway-SWMM software (hereinafter referred to as ISWMM), which is a city hydrology and rainwater and sewage drainage system simulation software developed by Beijing Inteli Environmental Technology Co., Ltd. It is a calculation platform that can simulate flexible hydraulic processes and is used to deduce runoff and external inflow through pipes, channels, water storage / treatment facilities and diversion structures, etc. The model includes the following functions: one-dimensional pipe network hydrodynamic process simulation, one-dimensional surface runoff production and concentration simulation, pollutant accumulation and flushing simulation, rainfall-induced inflow and infiltration simulation, snow simulation, groundwater simulation and external groundwater interaction. The one-dimensional pipe network hydrodynamic process simulation function module is used in the present application.

[0083] The basic data used to build ISWMM includes typical sewage discharge time series characteristic data of different land use types (residential land, commercial land, industrial land, public service land) in the study area, water structure data, drainage zoning data, population statistics data, elevation data, underlying surface attribute data, sub-catchment division data, meteorological data and calibration data. The typical sewage discharge time series characteristic data of different land use types in the study area is obtained by preferentially using drainage user water meters or sewage treatment plant monitoring data to obtain hourly sewage discharge time series characteristics of typical land use types. In the absence of measured data, it can be estimated through relevant literature or through existing statistical results of similar land use types. The water structure data includes the layout, structural parameters and operation logic data of facilities such as pump stations, gates, storage tanks, overflow outlets and diverters; the drainage zoning data includes drainage zoning, catchment range, control unit division data, pipe network data including pipe shape, pipe drainage type, pipe size, pipe type, node type, node internal bottom elevation, node depth and topological connection relationship of pipe network system, underlying surface attribute data including soil type, vegetation distribution, geological structure and land use type, sub-catchment division data being a standard vectorized shapfile format file obtained by dividing sub-catchment based on elevation data and drainage zoning data, meteorological data including rainfall time series data, temperature time series data, evaporation time series data; the calibration data includes historical monitoring data of flow, liquid level and water quality at monitoring points in the pipe network system.

[0084] In the present application, the data input process of ISWMM is as follows: in the input meteorological data, the rainfall time series data comes from the monitoring data of the rain gauge station, and the latitude and longitude coordinates of the rain gauge station need to be input in the present application. The latitude and longitude coordinates are used to match the positions of the sub-catchment and the rain gauge station. The matching method can adopt the nearest distance method or the inverse distance weighted method, so as to associate each sub-catchment to the most reasonable rainfall data source, and realize the accurate matching of the rainfall time series data and the sub-catchment. The model boundary data is obtained by studying the typical sewage discharge time sequence characteristic data of different land use types (residential land, commercial land, industrial land and public service land) in the research area, drainage zoning data, population statistics data and pipe network data, and finally a pipe network sewage inflow boundary data file is generated for input.

[0085] The specific steps for generating the pipe network sewage inflow boundary data file are as follows:

[0086] The land use layer and the drainage zoning data are spatially overlaid, and the population of each type of land (such as residential, commercial, industrial and public service land) in each drainage area is counted.

[0087] Combined with the typical sewage discharge time sequence characteristic curve of each type of land, the sewage inflow intensity of each drainage area is calculated on an hourly scale.

[0088] The sewage production of per capita is further refined through the population statistics data, and the total sewage discharge of each drainage area at different times is calculated.

[0089] The spatial mapping relationship between the drainage area and the actual pipe network node is established, the specific node position corresponding to the sewage inflow is determined, the calculated sewage quantity and water quality load are distributed to the corresponding node in the pipe network system, and the reasonable distribution of sewage inflow quantity and water quality is ensured.

[0090] The processed node sewage inflow distribution information is arranged into a pipe network sewage inflow boundary data file that meets the requirements of the ISWMM platform, and the accurate control of the space-time dynamic inflow process in the drainage system is realized.

[0091] Based on elevation and underlying surface attribute data, each sub-catchment is assigned corresponding attribute values. The nearest meteorological station's rainfall time series is extracted according to the geographical range of each sub-catchment. Specifically, the sub-catchments are matched with rain gauges using latitude and longitude coordinates to achieve the allocation and spatial linking of sub-catchments with rainfall time series data. After the drainage network model is constructed, meteorological data and sewage discharge time series characteristic data drive the ISWMM (Information and Water Quality Model). The model response is then tested to verify its rationality and ensure that the ISWMM can accurately represent and operate normally based on existing data. Using the above input data to drive the ISWMM, the ISWMM outputs initial simulation results, which are the simulation results of flow rate, liquid level, and water quality for each pipe segment. The simulation results of flow rate, liquid level, and water quality for each pipe segment are compared with historical monitoring data for the same parameters. If the error exceeds a set threshold (e.g., 30%), the ISWMM model parameters are adjusted and the simulation is repeated (including hydraulic parameters and water quality parameters) until the error does not exceed the set threshold.

[0092] After completing the calibration of ISSIM and ISWMM, the target surface water-groundwater coupling model and the target drainage network model are obtained, respectively. Inputting basic data, the target surface water-groundwater coupling model can output groundwater level time series data for each triangular grid cell within the target area, while the target drainage network model can output flow rate time series data, water quality time series data, and liquid level time series data for each pipe segment within the target area. In this invention, the flow rate time series data and water quality time series data are only used for ISWMM calibration.

[0093] S2. Using the spatial join function of GIS spatial analysis tools, establish a one-to-one spatial relationship mapping between each triangular grid unit of the target surface water-groundwater coupling model and each pipe segment of the target drainage network model.

[0094] Using GIS spatial analysis tools, establish the spatial correspondence between the triangular grid cells of the target surface water-groundwater coupled model and the pipe segments of the target drainage network model. Please refer to [reference needed]. Figure 2 As shown, Figure 2 Each small triangle represents a triangular grid cell, and each arrowed line segment represents a pipe segment. Using the above spatial correspondence, the drainage pipeline can be located to the groundwater simulation area where it is located, realizing a one-to-one spatial mapping of the simulation results of the two models.

[0095] S3. Using the basic data to drive the operation of the target surface water-groundwater coupled model and the target drainage network model, the groundwater level time series data of each triangular grid cell and the liquid level time series data of each pipe segment are obtained respectively.

[0096] To obtain the potential infiltration driving force of groundwater to the drainage pipe network, it is necessary to first build the pipe network operation benchmark state under the no-infiltration condition. The groundwater level time series data of each triangular grid element in the target area and the corresponding pipe segment liquid level time series data are compared, and the specific steps are as follows:

[0097] In the target drainage pipe network model, the groundwater simulation module is closed, and the meteorological data and typical sewage discharge time sequence characteristic data are used to drive the target drainage pipe network model to run, so as to obtain the liquid level time series data of each pipe segment in the target area, and obtain the reference condition without infiltration.

[0098] Under the same boundary conditions, the meteorological data are used to drive the target surface water-groundwater coupled model to run in parallel, so as to obtain the groundwater level time series data of the triangular grid element corresponding to each pipe segment.

[0099] S4, using the spatial relationship mapping established in step S2, the hourly liquid level difference time series data of the groundwater level and the pipe segment liquid level are calculated through the groundwater level time series data of each triangular grid element and the liquid level time series data of the corresponding pipe segment, and the risk index of the pipe segment is constructed using the hourly liquid level difference time series data.

[0100] Since the spatial corresponding relationship between each triangular grid element of the target surface water-groundwater coupled model and each pipe segment of the target drainage pipe network model has been established, the groundwater level data of the corresponding triangular grid element of each pipe segment and the pipe segment liquid level data are selected, and the comparison is made at the same time step to calculate the hourly liquid level difference time series.

[0101] Based on the hourly liquid level difference time series data, the liquid level difference threshold value for judging the existence of potential infiltration risk is set, and the risk index is constructed, including the average water level difference, the time proportion of exceeding the standard, and the cumulative excess water head.

[0102] The liquid level difference threshold value for judging the existence of potential infiltration risk can be set or dynamically calibrated according to the field monitoring data, historical case data backtracking or engineering experience:

[0103] Engineering experience: commonly used value range: 0.05m-0.20m; reference literature or existing urban drainage system risk criterion experience.

[0104] Historical case data backtracking: backtracking of existing historical infiltration cases or monitoring data; find out the liquid level difference data at the corresponding space-time node, take the average value or the conservative value to set the liquid level difference threshold value for judging the existence of potential infiltration risk.

[0105] Field monitoring data: typical well points are selected to lay underground water monitoring wells and conduct pipe network liquid level monitoring; the liquid level difference when the infiltration occurs is measured as the initial value of the liquid level difference threshold value for judging the existence of potential infiltration risk; and the threshold initial value is optimized and iteratively updated with the operation of the risk identification system.

[0106] The average water level difference is used to measure the average degree of the overall groundwater being higher than the potential infiltration threshold value of the pipe network, and only the positive values of the liquid level difference are averaged. This index is used to reflect the intensity of the pipe being in the state of being driven by groundwater for a long time, and the calculation formula is:

[0107]

[0108]

[0109] In the formula: is the average water level difference, with a unit of meters, n is the total number of times that meet t represents the tth time in the simulation period, T is the total number of times in the simulation period, is the liquid level difference at the tth time, with a unit of meters, is the liquid level difference threshold value for judging the existence of potential infiltration risk, with a unit of meters, is the groundwater level at the tth time, with a unit of meters, is the pipe segment liquid level at the tth time, with a unit of meters;

[0110] The exceeding time ratio is used to measure the proportion of the duration of groundwater being higher than the liquid level difference threshold value for judging the existence of potential infiltration risk in the total simulation period, and the calculation formula is:

[0111]

[0112]

[0113] In the formula: is the exceeding time ratio, is an indicator function, reflecting the persistence and frequency of potential infiltration driving force, converting the exceeding state of each time point t into binary value 1 or 0, which is convenient for statistics; the exceeding time ratio reflects the proportion of the time that the groundwater level is higher than the pipe network for a long time, and the value range is [0, 1]. That is indicates that the groundwater has never been higher than the pipe network; indicates that the groundwater has always been higher than the pipe network; The closer to 1, the more persistent and frequent the potential infiltration driving force of the groundwater on the pipe network.

[0114] Cumulative super-high water head estimates the potential infiltration scale by accumulating all super-high water head differences higher than the threshold of the liquid level difference that determines the potential infiltration risk, which can be used as a reliable alternative parameter of the potential infiltration scale (volume or total driving force) by quantifying the cumulative water head pressure of groundwater exceeding the pipe network, providing a scientific basis for identifying potential infiltration risk areas, and the calculation formula is:

[0115]

[0116]

[0117] In the formula: is the cumulative super-high water head, the unit is meter, is the time step, the unit is hour.

[0118] Cumulative super-high water head The greater the value, the higher the potential infiltration risk of groundwater to the pipe network, which may correspond to a larger total infiltration or a more serious area.

[0119] The above risk index setting has the advantages of quantification, visualization and integration, can be directly embedded into the calculation platform to realize the automatic identification and priority sorting of the potential infiltration risk area of the drainage pipe network, and support the subsequent risk warning, operation scheduling or reconstruction decision.

[0120] S5, set key control factors, and construct an uncertainty scenario sample set based on the disturbance of the key control factors.

[0121] In order to depict the uncertainty source of the potential infiltration risk of groundwater, the present application constructs a representative scenario sample set based on the disturbance of the control factors, constructs a scenario uncertainty analysis framework of the potential infiltration risk of groundwater, and uses it as the driving input of the model calculation.

[0122] S51, select target surface water-groundwater coupling model parameters, rainfall boundary factors and sewage boundary factors as key control factors, the target surface water-groundwater coupling model parameters include permeability coefficient, porosity, recharge rate and discharge coefficient, the rainfall boundary factors include storm recurrence period, rainfall duration and intensity-duration-frequency curve (hereinafter referred to as IDF) parameters, and the sewage boundary factors include unit population sewage intensity and sewage daily variation law;

[0123] S52, set the corresponding disturbance mode for different key control factors.

[0124] Targeted surface water-groundwater coupling model parameter disturbance mode: for each parameter, if the parameter has engineering measured or monitored data in the study area, the distribution fitting method (such as normal distribution, uniform distribution, etc.) is used to determine the disturbance interval and the probability distribution form; if the parameter has no direct data support, set a reasonable disturbance range according to the experience data or literature research, usually use uniform distribution or triangular distribution modeling, and set the relative fluctuation range of ±20%-±50%.

[0125] Rainfall boundary factor disturbance mode: according to the statistical characteristics of the target area historical meteorological data (such as rainfall time series, storm process data), combined with the target area historical storm measured data for statistical modeling, the control parameters in the design rain type are referenced to the target area storm measured process, and the rainfall process is disturbed according to the experience, and a reasonable offset interval is set, such as ±10%-±30%.

[0126] Sewage boundary factor disturbance mode: the unit population sewage intensity is disturbed by normal distribution or uniform distribution, and the daily variation factor sequence is randomly selected from the typical daily variation curve sample set to generate several representative daily variation factor sequences. Daily variation factor sequence refers to the time sequence variation proportion sequence of sewage flow or load in unit daily scale, which is usually normalized extracted from historical monitoring data. In the simulation, it is randomly called according to the scene to realize the diversification expression of sewage boundary condition.

[0127] S53, set the disturbance interval of different key control factors.

[0128] Targeted surface water-groundwater coupling model parameter disturbance interval:

[0129] If supported by measured or monitored data, the statistical value of the parameter is preferred to set the disturbance interval as:

[0130]

[0131]

[0132] In the formula: is the parameter the average, median or typical value in the monitoring period; the disturbance amplitude (for example 20%, 30%, 50%) of the parameter , which can be obtained by the following path: maximum and minimum value of historical monitoring value; statistical range of similar area sample; sensitivity analysis result in literature and expert experience judgment.

[0133] Rainfall boundary factor disturbance interval:

[0134] Rainstorm return period (hereinafter referred to as T) disturbance: Select multiple representative return periods (such as 2 years, 5 years, 10 years, 20 years, 50 years) to form a discrete disturbance sample set.

[0135] Rainfall duration disturbance (hereinafter referred to as D): Select multiple representative durations as discrete samples. Common durations include 5 minutes, 15 minutes, 30 minutes, 1 hour, 3 hours, and 6 hours. Ensure that typical scenarios of short heavy rain and long light rain are covered.

[0136] IDF parameter disturbance: For IDF parameters, use the IDF empirical formula fitted from regional measured rainfall data:

[0137]

[0138] In the formula: is the rainfall intensity under a certain rainstorm return period T and rainfall duration D; a, b, c, and d are fitting parameters.

[0139] Identify fitting parameters a, b, c, and d on the basis of fitting results from references or measured data, and set a disturbance range for each fitting parameter:

[0140]

[0141]

[0142]

[0143]

[0144] In the formula: * represents the reference value of the fitting parameter (fitting result from references or measured data), represents the disturbance amplitude (usually ±10%-30%), which is determined from historical fitting, regional research, or empirical data.

[0145] Sewage discharge boundary factor disturbance interval:

[0146] Per capita sewage discharge intensity disturbance: Combine sewage treatment plant inflow concentration monitoring data or statistical water discharge coefficients to set normal distribution or uniform distribution, and construct a disturbance sample set within ±30%;

[0147] Sewage discharge daily variation law disturbance: Obtain multiple typical daily variation curve samples (such as weekday type, weekend type, holiday type, extreme peak type, etc.), each curve corresponding to the discharge proportion of each time period within 24 hours. In the simulation process, randomly select or specify a curve as the hourly proportion input according to the scenario. Each curve sample can be obtained based on regional historical monitoring data statistics or set based on different types of life / sewage behavior, ensuring its normalization (the sum of 24-hour factors is 1) and physical rationality.

[0148] S54, a disturbance is applied to the key control factor to generate a key control factor disturbance sample, physical reasonableness checking and parameter constraint relationship checking are performed on the key control factor disturbance sample, and after unqualified samples are removed, an initial disturbance sample set of different key control factors is obtained.

[0149] Physical reasonableness checking: excluding parameters with unreasonable physical meaning (such as negative permeability), verifying each item by adopting a judgment rule of setting a range for each type of parameter, and judging and removing illegal values line by line. The typical rules adopted in the present application are as follows:

[0150] Target surface water-groundwater coupling model parameters:

[0151] Permeability coefficient ; porosity ; recharge rate , which cannot exceed a reasonable proportion (such as 70%) of the total annual rainfall; drainage coefficient .

[0152] Rainfall boundary factor:

[0153] Rainfall return period ; in the IDF parameter: , , , ,

[0154] Sewage boundary factor:

[0155] The sewage intensity per unit population is greater than 0, the sum of the daily variation factors is equal to 1, and the value of each hour is in the interval [0, 1].

[0156] Parameter constraint relationship checking: constraint judgment is performed on the associated parameters, a Boolean expression filter check_param_constraints(sample) is created for each type of parameter constraint to form a filter control condition set. The typical rules adopted in the present application are as follows:

[0157] Target surface water-groundwater coupling model parameters:

[0158] Porosity relationship constraint:

[0159]

[0160] In the formula: is the total porosity, is the effective porosity.

[0161] Recharge rate and infiltration amount relationship:

[0162]

[0163] wherein: is the recharge rate, is the infiltration amount.

[0164] Interlayer water conductivity ratio constraint:

[0165]

[0166] wherein: is the vertical hydraulic conductivity, is the horizontal hydraulic conductivity.

[0167] Hydraulic gradient constraint:

[0168]

[0169] wherein: is the water head difference, is the distance difference.

[0170] Rainfall boundary factor:

[0171] IDF parameter curve monotonically decreasing (i.e., rainfall intensity decreases with time D):

[0172]

[0173] Sewage boundary factor:

[0174] The fluctuation range of sewage load in different periods should be controlled in a reasonable range (to avoid the peak exceeding the actual capacity); the maximum fluctuation ratio of the normalized daily variation curve should not exceed a certain threshold (e.g., less than or equal to 3 times the average value).

[0175] S55, sampling and combining the initial perturbation sample set of each key control factor to construct an uncertainty scenario sample set.

[0176] This step includes the following sub-steps:

[0177] S551, Monte Carlo high-density sampling is performed on the initial perturbation sample set of the target surface water-groundwater coupling model parameters to obtain an initial parameter sample set. The initial parameter samples are preprocessed by normalization, and then a clustering algorithm is used to cluster the initial parameter sample set. From the clustering results, a number of samples of each category are extracted, the samples are de-normalized, and the de-normalized samples are used to construct a perturbation sample set of the target surface water-groundwater coupling model parameters.

[0178] Monte Carlo high-density sampling:

[0179] In the set parameter perturbation space, each parameter is independently randomly sampled to form an initial high-dimensional parameter sample set The sample size N is usually 500-5000; each sample point is a vector with dimension d (d represents the number of perturbation parameters), representing a complete set of model input parameter combinations.

[0180] Parameter space normalization:

[0181] After the generation of parameter perturbation samples, the original parameter data needs to be processed for scale unification. Because each physical parameter has different physical meaning and order of magnitude (such as the permeability coefficient is usually 10 −6 ~ 10 −4 m / s, and the porosity is usually a dimensionless value of 0.2-0.5), if not processed, some high-dimensional parameters will have a dominant influence on distance measurement in cluster analysis, affecting the accuracy of representative sample selection. To improve the performance of the clustering algorithm in high-dimensional parameter space, the parameter samples need to be normalized for preprocessing, so that each parameter falls within a unified scale range, avoiding the dominance of the absolute dimension of a parameter in the clustering results.

[0182] Normalization method:

[0183] For parameters with known physical range and relatively uniform distribution, each dimension of parameter x is normalized according to the following formula:

[0184]

[0185] In the formula: is the normalized parameter, is the parameter to be processed, is the minimum value of the parameter sample, is the maximum value of the parameter sample.

[0186] After processing, all parameters fall within the [0, 1] interval, which is conducive to the uniform weight processing of Euclidean distance; if there are out-of-bound samples (such as sampling beyond the empirical interval), they can be uniformly cropped or extrapolated.

[0187] For parameters that follow normal distribution or approximately symmetric distribution, Z-score method is used for normalization:

[0188]

[0189] In the formula: is the sample mean, is the sample standard deviation.

[0190] After processing, all parameters have a mean of 0 and a standard deviation of 1, which is conducive to preserving relative variability; it is suitable for cases where there are outliers or distribution fluctuations.

[0191] The specific processing method of the target surface water-groundwater coupling model parameters is shown in Table 1.

[0192] Table 1

[0193] Parameter type Processing method Explanation Permeability coefficient, excretion coefficient Z-score method With order of magnitude fluctuations, avoid clustering dominance Porosity Normalization Known physical range, suitable for scaling processing Infiltration rate Determined according to the distribution When the distribution is metamorphosed, consider Box-Cox transformation and then normalization processing

[0194] The processed sample set is saved in a unified format matrix for subsequent clustering algorithm calls. The dimensionality reduction projection result is verified by parameter principal component analysis (PCA) to determine whether there is abnormal deviation or excessive concentration; if the parameters are still highly correlated after normalization, feature dimensionality reduction or covariance filtering will be performed before clustering.

[0195] The clustering algorithm is used to perform spatial clustering on the initial parameter sample set, compressing redundant information and extracting representative parameter scenario clustering centers.

[0196] The optional clustering methods are as follows:

[0197] K-Means clustering: suitable for scenarios with a large number of samples (> 500), and easy to control the number of clusters K;

[0198] K-Medoids (PAM) clustering: more robust, with actual samples as clustering centers;

[0199] DBSCAN / hierarchical clustering: used to discover non-spherical clusters, suitable for complex distribution scenarios;

[0200] The number of clusters K can be automatically optimized according to the silhouette score, Davies-Bouldin index, etc., or empirically set to 5-10% of the total number of samples.

[0201] After clustering, one or more representative samples are extracted from each class of the clustering results as subsequent simulation samples (such as selecting class centers, the most dense samples within the class, etc.), and a target surface water-groundwater coupling model parameter perturbation sample set is constructed , K is the total number of samples.

[0202] S552, Latin hypercube sampling method is used for the initial perturbation sample set of the rainfall boundary factor, and a rainfall boundary factor perturbation sample set is constructed.

[0203] On the basis of having set the perturbation mode and interval of each parameter, the Latin hypercube sampling (LHS) method is used to construct a joint perturbation sample set in a high-dimensional space. The rainfall process uses a six-factor combination perturbation mode (i.e., the six perturbation factors of return period T, rainfall duration D, a, b, c, and d), each parameter has correlation but is treated as an independent dimension in the uncertainty sampling process, and is jointly constructed into a six-dimensional perturbation parameter space.

[0204] After determining the reasonable value range and distribution assumption of parameters T, a, b, c, d, D (which can be uniform distribution or normal distribution, etc., set according to historical measured data or design specifications), six dimensions are divided into layers, and a sample point is randomly selected in each layer. By recombining the sampling results of the six parameter dimensions, N groups of different disturbance parameter combinations are generated to form a high-dimensional sampling matrix:

[0205]

[0206] Each group of parameter combinations The corresponding design rainfall intensity value is calculated by the IDF empirical formula:

[0207]

[0208] In order to construct a complete rainfall process curve (rainfall-time sequence) that can be used for model driving, the are divided into several time periods, and the period rainfall is allocated according to and the assumed storm type (such as positively skewed, moderately skewed, negatively skewed or bell-shaped distribution), and finally a complete rainfall boundary factor disturbance sample set is formed.

[0209] S553, for the initial disturbance sample set of the sewage discharge boundary factor, the daily total sewage quantity time series data is obtained according to the unit population sewage intensity, combined with several groups of representative daily variation factor sequences and corresponding daily total sewage quantity data, the daily sewage discharge time series is generated by hour-by-hour multiplication operation, and the sewage discharge boundary factor disturbance sample set of the sewage discharge boundary factor is constructed.

[0210] Based on long-term historical observation data, monitoring results and actual operation experience, the system extracts typical sewage discharge daily variation coefficient curves. The typical working condition categories covered include but are not limited to: working day sewage discharge rule, holiday sewage discharge rule, rain and storm conditions sewage discharge rule, drought or low discharge period rule. Each typical curve takes 24 hours as the time period, expresses the hourly proportion distribution of sewage discharge, meets the coefficient normalization requirement (24-hour coefficient sum is 1), and reflects the daily variation characteristics of sewage discharge intensity.

[0211] In order to reflect the fluctuation of total sewage discharge, each selected typical daily variation coefficient curve is simulated with the total sewage quantity of the day with random disturbance. The disturbance amplitude is usually set to ±20%, and uniform distribution or normal distribution random sampling is adopted to ensure the reasonableness and diversity of sewage discharge variation.

[0212] N representative typical curves are selected from a typical daily variation coefficient library, covering main working condition categories. For each typical curve, M different total sewage volume perturbation values are generated, forming the diversity of total discharge. Combined with the typical daily variation coefficient and the total sewage volume after perturbation, NXM sets of pollution discharge law scenarios are constructed, obtaining the boundary factor perturbation sample set of pollution discharge. Each pollution discharge law is obtained by multiplying the typical daily coefficient and the total sewage volume of the day, ensuring the typicality of the daily pollution discharge proportion, while introducing the dynamic change of the pollution discharge volume.

[0213] S554, the obtained target surface water-groundwater coupling model parameter perturbation sample set, rainfall boundary factor perturbation sample set and initial perturbation sample set of pollution discharge boundary factor are combined to obtain an uncertainty scenario sample set, and the expression of a sample in the uncertainty scenario sample set is:

[0214]

[0215] In the formula: is a certain sample in the uncertainty scenario sample set, is the i th sample in the target surface water-groundwater coupling model parameter perturbation sample set, is the j th sample in the rainfall boundary factor perturbation sample set, is the k th sample in the pollution discharge boundary factor perturbation sample set.

[0216] The sample in the uncertainty scenario sample set The input path is: and Input the target drainage pipe network model, and Input the target surface water-groundwater coupling model.

[0217] S6, the uncertainty scenario sample set is input into the target surface water-groundwater coupling model and the target drainage pipe network model respectively, to obtain the simulation calculation results of the groundwater level time series data and the pipe segment liquid level time series data, and the simulation calculation results are input into step S4 to calculate the risk index of each pipe segment under all uncertainty scenarios, and the potential infiltration risk area is determined according to the risk index.

[0218] The target surface water-groundwater coupling model and the target drainage pipe network model are driven by the uncertainty scenario sample set to obtain simulation calculation results, and the specific steps include:

[0219] The parameters of the uncertainty scenario sample set are automatically imported into the above model to realize automatic batch driving calculation of the model. This process is driven by an automated simulation scheduling script (Python batch processing tool) and is executed in parallel on a high-performance computing cluster (SLURM-based HPC platform). By standardizing the format, the parameters are automatically converted into the model input format to ensure consistency and accuracy of the parameters. The model batch is automatically executed, and the simulation calculation under all uncertainty scenarios is completed in sequence, and the corresponding simulation output results are saved.

[0220] The simulation calculation result data is extracted from the output of the target surface water-groundwater coupling model and the target drainage pipe network model, respectively. The former obtains groundwater level time series data, and the latter obtains pipe segment liquid level time series data. The extracted multiple uncertainty scenario result data is uniformly formatted and Z-score normalized to ensure consistent spatial and temporal resolution to meet the subsequent risk identification and analysis requirements.

[0221] Three risk indicators for each pipe segment under each uncertainty scenario are calculated, and through integrated analysis of the risk indicators under multiple scenarios, the identification and judgment of the potential infiltration risk area of groundwater are supported.

[0222] The steps for determining the potential infiltration risk area according to the risk indicators are as follows:

[0223] The risk indicators are normalized and a comprehensive risk indicator is set. The calculation formula of the comprehensive risk indicator is:

[0224]

[0225] In the formula: is the comprehensive risk indicator of the ith pipe segment under all uncertainty scenarios, , and is the set weight, and the equal weight allocation is used by default, is the normalized average water level difference of the ith pipe segment under all uncertainty scenarios, is the normalized over-standard time proportion of the ith pipe segment under all uncertainty scenarios, is the normalized cumulative over-high water head of the ith pipe segment under all uncertainty scenarios.

[0226] According to the comprehensive risk indicator results, the quantile method is used in combination with the set threshold based on expert experience to determine and divide the potential infiltration risk level of each pipe segment. The steps are as follows:

[0227] The comprehensive risk indicators of all pipe segments are collected to form a risk indicator dataset, and the quantile method is used to divide the risk indicator dataset into several quantile intervals (using quartiles as an example, as the first quartile, as the second quartile, as the third quartile, each quartile interval is mapped to a predetermined risk level:

[0228] corresponding to a low risk level;

[0229] corresponding to a low-medium risk level;

[0230] corresponding to a medium-high risk level;

[0231] corresponding to a high risk level.

[0232] After completing the basic level division, the upper and lower threshold values of the quartile interval obtained by the division are adjusted based on expert experience;

[0233] Using the adjusted quartile interval, the comprehensive risk index of each pipe segment is mapped to the corresponding risk level, and the potential infiltration risk level classification results of all pipe segments are obtained.

[0234] The method for adjusting the upper and lower threshold values of the quartile interval obtained by the division is as follows:

[0235] Set the experience threshold range: combine historical infiltration events, groundwater high water level period statistics and measured infiltration intensity to set the reference score boundary value, such as:

[0236] corresponding to a low risk level;

[0237] corresponding to a medium risk level;

[0238] corresponding to a high risk level.

[0239] Expert fine-tuning: organize experts to review the preliminary level results, and can make moderate fine-tuning to some quartile points (such as increasing the corresponding value from 0.81 to 0.85), to ensure that the grading standard is more in line with the actual regional management.

[0240] To enhance the empirical basis of the grading standard, the score boundary can also be calibrated by typical historical events: select typical groundwater infiltration events (such as overflow, flow surge, water level anomaly) and mark the affected pipe segments; extract the score of the affected pipe segments and compare it with the preset level threshold value to identify whether the score value is consistent with the severity of the event; combine ROC curve, confusion matrix and other indicators to optimize the threshold setting and improve the accuracy and verifiability of the grading discrimination.

[0241] ​To prove the feasibility and effectiveness of the above method, the following is illustrated in a specific implementation scenario. The present application selects the basic data of a target area in 2023, processes and calculates it through the above method, and obtains the corresponding test results.

[0242] Please refer to Figure 3 , it can be seen that Figure 3 is a groundwater level spatial distribution diagram of a local area at a certain time node in the target area. For the triangular grid elements divided in the target area, the redder the color, the higher the groundwater level, and the bluer the color, the lower the groundwater level.

[0243] Please refer to Figure 4 , it can be seen that Figure 4 is a groundwater depth and rainfall time series curve obtained by calculation in 2023. Groundwater depth is the vertical distance from the groundwater surface to the ground surface, which can intuitively reflect the change of groundwater level, Figure 4 The middle blue curve is the groundwater depth time series curve, and the red curve is the rainfall time series curve. From Figure 4 , it can be seen that under different rainfall events, the response of groundwater level is obviously different, and it can respond to rainfall events in real time.

[0244] Please refer to Figure 5 , it can be seen that Figure 5 is the classification and identification result of the potential infiltration risk area of the local area in the target area obtained by step S6 evaluation and calculation. The dots in the figure represent the corresponding pipe sections, dark purple represents high risk level corresponding to the pipe section, light purple represents medium risk level corresponding to the pipe section, and white represents low risk level corresponding to the pipe section. From Figure 5 , it can be seen that the main pipe and part of the branch pipe face high potential infiltration risk, and there are also cases where the potential infiltration risk level difference between two pipe sections with close geographical distance is very large, which is related to the height of the pipe network and the local height of the groundwater level. It is consistent with the actual measurement results.

Claims

1. An intelligent grading identification method for potential infiltration risk areas of a sewer network, characterized in that, The method comprises the following steps: S1, obtaining basic data of a target area, constructing and calibrating an initial surface water-groundwater coupling model and an initial drainage pipe network model respectively to obtain a target surface water-groundwater coupling model and a target drainage pipe network model; S2, using a GIS spatial analysis tool, establishing a corresponding or adjacent relationship between each triangular grid cell of the target surface water-groundwater coupling model and each pipe section of the target drainage pipe network model based on spatial positions; S3, closing a groundwater simulation module in the target drainage pipe network model, inputting meteorological data and typical sewage discharge time sequence characteristic data in the basic data to drive the target drainage pipe network model to run, and obtaining liquid level time sequence data of each pipe section in the target area; using the same meteorological data to drive the target surface water-groundwater coupling model to run in parallel, and obtaining groundwater level time sequence data of the triangular grid cells corresponding to the pipe sections; S4, using the corresponding or adjacent relationship between the triangular grid cells and the pipe sections, calculating liquid level difference data of each triangular grid cell by subtracting the liquid level of the corresponding pipe section from the groundwater level of the triangular grid cell, constructing hourly liquid level difference time sequence data, and constructing a risk index of each pipe section according to the liquid level difference time sequence data; S5, setting a plurality of key control factors, and constructing an uncertainty scenario sample set by perturbing the key control factors; S6, inputting the uncertainty scenario sample set into the target surface water-groundwater coupling model and the target drainage pipe network model respectively for simulation, obtaining groundwater level and pipe section liquid level time sequence results under each scenario; using the results for risk index calculation in S4 to obtain the risk index of each pipe section under all uncertainty scenarios, and determining the grading results of a potential infiltration risk area according to the risk index of each pipe section.

2. The sewer network potential infiltration risk area intelligent grading identification method according to claim 1, characterized in that: In the step S1, the basic data includes basic data of the initial surface water-groundwater coupling model and basic data of the initial drainage pipe network model; the basic data of the initial surface water-groundwater coupling model includes elevation data, a research range, a pipe network distribution range, underlying surface attribute data, meteorological data and calibration data; and the basic data of the initial drainage pipe network model includes typical sewage discharge time sequence characteristic data of different land use types in the research area, hydraulic structure data, drainage zoning data, pipe network data, population statistics data, elevation data, underlying surface attribute data, sub-catchment division data, meteorological data and calibration data.

3. The sewer network potential infiltration risk area intelligent grading identification method according to claim 2, characterized in that: In the step S1, the initial surface water-groundwater coupling model is constructed based on the inteliway-SSIM software, and the steps of constructing and calibrating the initial surface water-groundwater coupling model are as follows: According to the elevation data, the research range and the pipe network distribution range, the target area is divided into triangular grid cells in the inteliway-SSIM; Based on the elevation data and the underlying surface attribute data, each triangular grid cell is assigned with corresponding attribute parameters; inputting the meteorological data to drive the inteliway-SSIM to run, and obtaining a groundwater level data simulation result; The calibration data is historical monitoring data of groundwater level, and the simulation result of groundwater level is compared with the historical monitoring data of groundwater level. If the error exceeds the set threshold, the model parameters of inteliway-SSIM are adjusted, or the grid is re-divided and re-simulated until the error does not exceed the set threshold.

4. The sewer network potential infiltration risk area intelligent grading identification method according to claim 2, characterized in that: In step S1, the construction of the initial drainage pipe network model is based on inteliway-SWMM software, and the steps of constructing and calibrating the initial drainage pipe network model are as follows: According to the elevation data and the underlying surface attribute data, each sub-catchment obtained by division is given a corresponding attribute value, and the corresponding rainfall time series of the nearest weather station is extracted according to the geographical range of each sub-catchment; The meteorological data and typical sewage discharge time series characteristic data are input to drive inteliway-SWMM to run and obtain the simulation results of the flow, liquid level and water quality of each pipe section; The calibration data is the historical monitoring data of the flow, liquid level and water quality of each pipe section, and the simulation results of the flow, liquid level and water quality of each pipe section are compared with the historical monitoring data of the flow, liquid level and water quality of each pipe section. If the error exceeds the set threshold, the model parameters of inteliway-SWMM are adjusted and the simulation is repeated until the error does not exceed the set threshold.

5. The sewer network potential infiltration risk area intelligent grading identification method according to claim 1, characterized in that: In step S4, the steps of constructing the risk index of each pipe section are as follows: Based on the hourly liquid level difference time series data, the liquid level difference threshold for judging the existence of potential infiltration risk is set, and the risk index of the pipe section is constructed, including the average water level difference, the over-standard time proportion and the cumulative over-high water head; The average water level difference calculation formula is: ， ， In the formula: is the average water level difference, with units of meters, n is the total number of times that satisfy , t represents the tth time in the simulation period, T is the total number of times in the simulation period, is the liquid level difference at the tth time, with units of meters, is the liquid level difference threshold value for judging the existence of potential infiltration risk, with units of meters, is the groundwater level at the tth time, with units of meters, is the pipe segment liquid level at the tth time, with units of meters; The over-standard time proportion calculation formula is: ， ， In the formula: is the over-standard time ratio, is an indicator function; The cumulative over-high water head calculation formula is: ， ， wherein: is the cumulative super-high water head in meters, is the time step in hours.

6. The sewer network potential infiltration risk area intelligent grading identification method according to claim 1, characterized in that, The step S5 includes the following sub-steps: S51, select the target surface water-groundwater coupling model parameters, rainfall boundary factors and sewage boundary factors as key control factors, the target surface water-groundwater coupling model parameters include permeability coefficient, porosity, recharge rate and discharge coefficient, the rainfall boundary factors include storm recurrence period, rainfall duration and intensity-duration-frequency curve parameters, and the sewage boundary factors include unit population sewage intensity and sewage daily variation law; S52, set the corresponding disturbance mode for different key control factors; S53, set the disturbance interval of different key control factors according to historical monitoring data or expert knowledge; S54, apply disturbance to the key control factors to generate key control factor disturbance samples, and perform physical rationality check and parameter constraint relationship check on the samples, and after eliminating unqualified samples, the initial disturbance sample set of each key control factor is obtained; S55, sample and combine the initial disturbance sample set of each key control factor to construct an uncertainty scenario sample set.

7. The intelligent grading identification method for potential infiltration risk areas of drainage pipe network according to claim 6, characterized in that: The step S52 includes: The distribution fitting method is used to determine the disturbance interval and probability distribution form of the target surface water-groundwater coupling model parameters; The statistical characteristics of the rainfall boundary factor are inferred according to historical meteorological data of the target area, statistical modeling is performed in combination with historical storm rainfall data of the target area, and the rainfall process is empirically disturbed according to the storm classification standard; The unit population pollution intensity is disturbed by using a normal distribution or a uniform distribution, and the pollution daily variation law is generated by randomly selecting a sequence of representative daily variation factors from a sample set of typical daily variation curves.

8. The intelligent grading identification method of the potential infiltration risk area of the drainage pipe network according to claim 6, characterized in that: The step S55 comprises: S551, performing Monte Carlo high-density sampling on the initial disturbance sample set of the target surface water-groundwater coupling model parameters to obtain an initial parameter sample set, performing normalization preprocessing on the initial parameter sample, performing clustering on the initial parameter sample set by using a clustering algorithm, extracting a plurality of samples of each category from the clustering result, performing inverse normalization on the samples, and constructing a disturbance sample set of the target surface water-groundwater coupling model parameters by using the inverse normalized samples; S552, constructing a disturbance sample set of the rainfall boundary factor by using a Latin hypercube sampling method on the initial disturbance sample set of the rainfall boundary factor; S553, constructing a disturbance sample set of the pollution boundary factor by generating a daily sewage discharge time sequence through hour-by-hour multiplication operation of a plurality of groups of representative daily variation factor sequences and corresponding daily total sewage volume data according to the daily total sewage volume time series data obtained from the unit population pollution intensity, and combining the initial disturbance sample set of the pollution boundary factor; S554, combining the disturbance sample set of the target surface water-groundwater coupling model parameters, the disturbance sample set of the rainfall boundary factor, and the initial disturbance sample set of the pollution boundary factor to obtain an uncertainty scenario sample set, and the sample expression in the uncertainty scenario sample set is: ， wherein: is a certain sample in the uncertainty scenario sample set, is the i-th sample in the target surface water-groundwater coupling model parameter perturbation sample set, is the j-th sample in the rainfall boundary factor perturbation sample set, is the k-th sample in the pollution discharge boundary factor perturbation sample set.

9. The sewer network potential infiltration risk area intelligent grading identification method according to claim 5, characterized in that, In step S6, the step of determining the grading result of the potential infiltration risk area according to the risk index of each pipe section is as follows: performing normalization processing on the risk index, and setting a comprehensive risk index, and the calculation formula of the comprehensive risk index is: ， In the formula: is the comprehensive risk index of the i-th pipe section under all uncertainty scenarios, , and is the set weight, is the normalized average water level difference of the i-th pipe section under all uncertainty scenarios, is the normalized exceeding time ratio of the i-th pipe section under all uncertainty scenarios, is the normalized cumulative excess head of the i-th pipe section under all uncertainty scenarios. According to the comprehensive risk index result, the potential infiltration risk level of each pipe section is determined and divided by using a quantile method and combining a threshold value set based on expert experience.

10. The sewer network potential infiltration risk area intelligent grading identification method according to claim 9, characterized in that, The step of determining and dividing the potential infiltration risk level of each pipe section by using a quantile method and combining a threshold value set based on expert experience is as follows: All comprehensive risk indexes of all pipe sections under all uncertainty scenarios are collected to form a risk index data set, the data set is divided into a plurality of quantile intervals by using a quantile method, and each quantile interval is corresponded to a predetermined risk level; The upper and lower threshold values of the divided quantile intervals are adjusted based on expert experience; The comprehensive risk index of each pipe section is mapped to the corresponding risk level by using the adjusted quantile intervals to obtain the grading result of the potential infiltration risk level of all pipe sections.

Images