Urban intelligent drainage management system based on big data analysis

Abstract

The invention provides an urban intelligent drainage management system based on big data analysis, and relates to the technical field of urban drainage management, and the system comprises a multi-source data collection module which collects historical climate changes, rainfall curves, satellite meteorology, pipe network node water level flow and topological structure data in real time; the dynamic three-dimensional modeling module constructs a pipe network three-dimensional model according to a fluid mechanics principle, and binds real-time sensor data by a spatio-temporal data fusion algorithm; the drainage limit analysis module calculates the maximum drainage threshold value and the failure probability of the node under extreme rainfall; the intelligent early warning module fuses LSTM and ARIMA models, generates a drainage pressure thermodynamic diagram in the future 72 hours in combination with satellite meteorological data and triggers early warning; the health assessment module generates a node health index and a maintenance priority order; the resource scheduling engine generates a dynamic scheduling scheme. The urban drainage problem is effectively solved, the ability of a drainage system to deal with extreme weather and pipe network aging problems is improved, and accurate early warning and efficient resource allocation are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban drainage management, and in particular to an urban intelligent drainage management system based on big data analysis. Background Art

[0002] With the acceleration of urbanization, the continuous expansion of urban scale and the continuous growth of population density, urban drainage systems are facing unprecedented severe challenges. Traditional urban drainage systems have exposed many problems in dealing with increasingly complex drainage needs.

[0003] On the one hand, extreme weather events are becoming more frequent, with the intensity and frequency of rainstorms increasing significantly. Such extreme rainfall can easily lead to urban waterlogging, causing huge losses to urban infrastructure, residents' lives and economic development.

[0004] On the other hand, due to the long-term influence of factors such as sewage corrosion, mechanical wear and underground soil displacement, the structural integrity of the pipeline network is easily damaged, and problems such as pipe wall corrosion and sediment accumulation occur frequently, resulting in a significant decline in drainage capacity. The drainage pipeline network in some old urban areas even has serious leakage and blockage.

[0005] In addition, the existing drainage system management model is relatively extensive, lacking effective real-time monitoring and precise control methods. Key parameters such as water level and flow in the pipe network cannot be obtained in a timely and accurate manner, making it difficult to conduct a comprehensive and real-time assessment of the operating status of the drainage system. When faced with sudden drainage problems, management departments are often unable to make scientific decisions quickly, resulting in untimely problem handling, which further exacerbates the impact of disasters.

[0006] In order to effectively solve these problems and improve the safety, reliability and operating efficiency of urban drainage systems, it is imperative to develop an urban smart drainage management system based on big data analysis. The system can integrate multi-source data, use advanced algorithms and models, realize refined management and intelligent decision-making of urban drainage systems, and provide strong guarantees for the sustainable development of cities. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide an urban intelligent drainage management system based on big data analysis, which can improve fleet management efficiency and operational benefits.

[0008] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0009] An urban smart drainage management system based on big data analysis, comprising:

[0010] Multi-source data acquisition module, used to obtain historical climate change data, high-precision historical rainfall curves, satellite meteorological monitoring data, water level sensor data at each node of the drainage network, drainage flow sensor data and network topology data in real time;

[0011] The dynamic 3D modeling module builds a 3D model of the urban drainage network that includes network material properties, terrain slope, and node connection relationships based on the principles of fluid mechanics, and dynamically binds real-time sensor data to the 3D model of the urban drainage network through a data fusion algorithm;

[0012] The drainage limit analysis module uses the pressure-flow coupling algorithm to calculate the maximum drainage threshold of each node under extreme rainfall conditions, and generates a node failure probability distribution diagram based on Monte Carlo simulation;

[0013] The intelligent early warning module generates a heat map of drainage pressure in the next 72 hours by integrating the long short-term memory network LSTM and the ARIMA time series model, combined with satellite meteorological forecast data. When the real-time data of the node exceeds the preset safety threshold, a multi-level early warning mechanism is triggered;

[0014] The health assessment module uses the random forest algorithm to analyze the correlation between the node's historical fault data and real-time performance indicators, generates a node health index including the pipe wall corrosion rate and sediment accumulation coefficient, and outputs the maintenance priority ranking;

[0015] The resource scheduling engine optimizes maintenance path planning based on genetic algorithms, generates dynamic scheduling plans based on comprehensive warning levels, health indexes, and emergency resource distribution, and achieves real-time optimal configuration of personnel, equipment, and materials.

[0016] Furthermore, the multi-source data acquisition module includes:

[0017] Multi-parameter sensor arrays deployed at key nodes of the drainage network, including ultrasonic flow meters, laser turbidity meters, and pressure transmitters;

[0018] Meteorological satellite data interface, supporting the reception of high-resolution synthetic aperture radar SAR rainfall monitoring data;

[0019] The municipal geographic information system GIS data fusion unit realizes the three-dimensional coordinate calibration of the pipe network topology structure.

[0020] Furthermore, the dynamic 3D modeling module adopts a spatiotemporal data fusion algorithm, and its fusion weight coefficient is calculated in the following way:

[0021] Sensor data contribution part: multiply the real-time reading of each sensor by the geographic influence factor and sum it, then divide it by the maximum reading of the sensor for normalization, and multiply it by the dynamic adjustment parameter α;

[0022] Satellite data contribution: The rainfall forecast value of each satellite meteorological channel is multiplied by the time attenuation coefficient and then summed, divided by the maximum value of the satellite data for normalization, and multiplied by the dynamic adjustment parameter β;

[0023] The final fusion weight coefficient is the weighted sum of the above two parts, where α+β=1 and is updated every 5 minutes;

[0024] The geographical impact factor is calculated based on the node altitude and surrounding building density, with a value range of 0.8-1.2;

[0025] The time attenuation coefficient decays dynamically according to an exponential function, expressed as 0.5 multiplied by e to the power of -0.1t.

[0026] Furthermore, the failure probability calculation method of the drainage limit analysis module is:

[0027] Conduct at least 10,000 Monte Carlo simulations, and calculate the actual drainage flow of the node in each simulation;

[0028] The actual discharge is compared to a threshold value obtained by subtracting a correction term for historical bias from the theoretical maximum discharge;

[0029] The proportion of the simulation times when the actual traffic exceeds the threshold to the total times is counted as the node failure probability;

[0030] Correction mechanism: The historical deviation correction term is the multiplication of the historical maximum flow fluctuation value and the dynamic adjustment factor γ. γ is adaptively adjusted through the hyperbolic tangent function, and the adjustment range is affected by the difference between the current flow fluctuation and the standard deviation of the historical data.

[0031] Furthermore, the hybrid prediction model of the intelligent early warning module performs the following operations:

[0032] Prediction formula logic:

[0033] LSTM prediction component: Generates prediction values ​​based on drainage pressure data in a historical time window (24-72 hours);

[0034] ARIMA forecast component: Generates benchmark forecast values ​​through an autoregressive difference moving average model;

[0035] Satellite correction component: the impact of future rainfall intensity predicted by superimposed satellite on drainage pressure;

[0036] Weight distribution rules:

[0037] The weights of LSTM and ARIMA are dynamically calculated by the ratio of their prediction errors and normalized using the sigmoid function;

[0038] The meteorological sensitivity coefficient controls the contribution intensity of the satellite correction term and has a value range of 0.2-0.8.

[0039] Furthermore, the health index calculation rule of the health assessment module is:

[0040] Calculate the ratio of the current value to the maximum allowable value for indicators such as corrosion depth and deposition thickness;

[0041] A dynamic weight is assigned to each indicator, which is calculated by the entropy weight method and the sum is 1;

[0042] Multiply the score of each indicator by the exponential decay factor. The decay coefficient λ reflects the aging rate of the pipeline network. The value range of λ is 0.05-0.2 / year.

[0043] The final health index is the standardized value of the weighted sum of all indicator scores.

[0044] Furthermore, the optimization objective function of the resource scheduling engine is defined as:

[0045] Total path time: the sum of the path distances of each rescue point divided by the dynamic driving speed;

[0046] Time breach cost: impose a penalty on the response time that exceeds the time limit of the plan, and the penalty coefficient η is positively correlated with the warning level;

[0047] Total resource constraints: The available resources must meet the demand threshold;

[0048] Decision variable constraints: Resource allocation adopts 0-1 integer programming.

[0049] Furthermore, the geographic impact factor is calculated using the following rules:

[0050] Altitude influence term: the normalized value of the node altitude relative to the regional extreme value multiplied by 0.2;

[0051] Building density impact term: 1 minus the ratio of the building density around the node to the maximum density of the city, multiplied by 0.8;

[0052] The final geographic impact factor is the sum of the above two items.

[0053] Furthermore, the update rule of the dynamic adjustment factor γ is:

[0054] Based on the benchmark value of 0.8;

[0055] The adjustment range is calculated by the difference between the current flow fluctuation and the historical fluctuation through the hyperbolic tangent function;

[0056] Adjust the amplitude coefficient μ to control the range of change, with a value of 0.2-0.6.

[0057] Furthermore, the optimization based on the genetic algorithm performs the following steps:

[0058] The non-dominated sorting genetic algorithm NSGA-II is used to generate the Pareto optimal solution set;

[0059] Define the dual objective vector: total path time and time breach cost;

[0060] The optimal solution is selected according to the real-time weather warning level, and the time default cost is minimized first when the red warning occurs.

[0061] The above solution of the present invention includes at least the following beneficial effects:

[0062] Accurate data collection and analysis: The multi-source data collection module comprehensively collects various data such as meteorology, pipe network status and topological structure, providing a solid data foundation for subsequent modules; the dynamic 3D modeling module builds an accurate 3D model based on this, and updates it in real time through the spatiotemporal data fusion algorithm, which can accurately reflect the actual operation status of the drainage pipe network and help more accurate analysis and decision-making; for example, when predicting extreme situations such as floods, it can provide accurate pipe network water flow simulation, providing strong support for early prevention;

[0063] Efficient drainage limit assessment: The drainage limit analysis module uses the pressure-flow coupling algorithm and the improved Monte Carlo simulation algorithm to accurately calculate the maximum drainage threshold and failure probability of each node under extreme rainfall. The generated node failure probability distribution map allows managers to intuitively understand the weak links of the pipeline network and formulate response measures in advance to reduce the risk of disasters such as waterlogging;

[0064] Timely and accurate warning: The intelligent warning module integrates LSTM and ARIMA models and combines satellite meteorological data to accurately predict the drainage pressure in the next 72 hours. The generated heat map intuitively displays the pressure distribution. When the node data is abnormal, the multi-level warning mechanism responds quickly, allowing the management department to prepare in advance and reduce disaster losses.

[0065] Scientific health assessment: The health assessment module uses a multi-feature fusion assessment algorithm to comprehensively consider real-time performance indicators, historical fault data, and aging attenuation effects, accurately calculate and sort the node health index; help management departments to reasonably arrange maintenance plans, give priority to handling nodes with serious problems, improve maintenance efficiency, and extend the service life of the pipeline network;

[0066] Optimize resource scheduling: The resource scheduling engine integrates the warning level, node health index and emergency resource distribution, and uses multi-objective optimization algorithms and genetic algorithms to achieve optimal resource allocation; generate dynamic scheduling plans to ensure that in an emergency, personnel, equipment and materials can quickly reach the fault point, improve emergency response capabilities, and ensure the stable operation of the urban drainage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic diagram of an urban intelligent drainage management system based on big data analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0069] like Figure 1 As shown, an embodiment of the present invention proposes an urban intelligent drainage management system based on big data analysis, including a multi-source data acquisition module, a dynamic three-dimensional modeling module, a drainage limit analysis module, an intelligent early warning module, a health assessment module and a resource scheduling engine.

[0070] In this embodiment, the multi-source data acquisition module is used to obtain historical climate change data, high-precision historical rainfall curves, satellite meteorological monitoring data, water level sensor data of each node of the drainage network, drainage flow sensor data and network topology data in real time;

[0071] Furthermore, the multi-source data acquisition module includes:

[0072] Multi-parameter sensor arrays deployed at key nodes of the drainage network, including ultrasonic flow meters, laser turbidity meters, and pressure transmitters;

[0073] Meteorological satellite data interface, supporting the reception of high-resolution synthetic aperture radar SAR rainfall monitoring data;

[0074] Municipal Geographic Information System GIS data fusion unit to achieve three-dimensional coordinate calibration of the pipe network topology structure;

[0075] The multi-source data acquisition module is a basic component of the urban smart drainage management system based on big data analysis. Its main function is to obtain all kinds of data related to urban drainage in real time and comprehensively, and provide data support for subsequent analysis, modeling and decision-making. The following is a detailed description of the module:

[0076] Overall function overview:

[0077] The multi-source data acquisition module is responsible for collecting data from a variety of different sources. These data cover multiple aspects such as meteorology, drainage network status, and network topology, which helps the system to fully understand the current status and potential influencing factors of urban drainage;

[0078] Specific data collection content:

[0079] Meteorological data:

[0080] Historical climate change data: collect long-term climate records of the city and surrounding areas, including information such as temperature, humidity, air pressure, wind direction, and wind speed; these data can help analyze the long-term impact of climate change on the city's drainage system. For example, temperature changes may affect the rate of ice and snow melting, which in turn affects drainage flow;

[0081] High-precision historical rainfall curve: records information such as rainfall intensity, duration, and rainfall distribution in different time periods in the past; high-precision rainfall curves can provide an important basis for predicting future rainfall conditions and evaluating the performance of drainage systems under different rainfall conditions;

[0082] Satellite meteorological monitoring data: Receive high-resolution synthetic aperture radar (SAR) rainfall monitoring data through the meteorological satellite data interface; satellite data can provide large-scale, real-time meteorological information, which helps to timely grasp the spatial distribution and change trend of rainfall, especially for the monitoring and early warning of sudden heavy rainfall events.

[0083] Drainage network status data:

[0084] Water level sensor data at each node of the drainage network: Water level sensors are deployed at key nodes of the drainage network to monitor the water level at the nodes in real time. Water level data can reflect the drainage capacity and operation status of the network. When the water level exceeds a certain threshold, it may indicate poor drainage or blockage.

[0085] Drainage flow sensor data: Install drainage flow sensors to measure the water flow at each node in the pipe network; flow data can help evaluate the actual drainage load of the pipe network, and combined with water level data, it can more accurately determine the operation status of the pipe network;

[0086] Pipeline network topology data:

[0087] Through the municipal geographic information system (GIS) data fusion unit, the topological structure data of the drainage network is obtained, including the layout, pipe diameter, pipe material, node connection relationship and other information of the network; these data are the basis for building a three-dimensional model of the urban drainage network, which helps to accurately simulate the flow of water in the network;

[0088] Data acquisition equipment and interface:

[0089] Multi-parameter sensor array:

[0090] Deploy multi-parameter sensor arrays at key nodes of the drainage network, including ultrasonic flow meters, laser turbidity meters, and pressure transmitters;

[0091] Ultrasonic flow meter: It uses the characteristics of ultrasonic waves propagating in fluids to measure the flow velocity and flow rate of fluids. It has the advantages of non-contact measurement, high accuracy, and strong reliability.

[0092] Laser turbidity meter: measures the concentration of suspended particles in water through the principle of laser scattering, reflecting the turbidity of water quality; turbidity data can help determine whether there are problems such as sediment accumulation or sewage leakage in the pipe network;

[0093] Pressure transmitter: measures the pressure changes in the pipe network. The pressure data can reflect the operating pressure status of the pipe network and plays an important role in detecting leakage, blockage and other faults in the pipe network.

[0094] Meteorological satellite data interface:

[0095] A meteorological satellite data interface that supports receiving high-resolution synthetic aperture radar (SAR) rainfall monitoring data; this interface can realize real-time data transmission with the meteorological satellite system, ensuring that the system can obtain the latest meteorological information in a timely manner;

[0096] Municipal Geographic Information System (GIS) Data Fusion Unit:

[0097] Realize the three-dimensional coordinate calibration of the topological structure of the pipe network; through data fusion with the GIS system, combine the topological structure data of the drainage pipe network with geographic spatial information to improve the accuracy and visualization of the data, and provide more accurate basic data for subsequent analysis and modeling;

[0098] Real-time and accuracy of data collection:

[0099] The multi-source data acquisition module emphasizes real-time and accuracy. It uses advanced sensor technology and data transmission technology to ensure timely and accurate acquisition of various types of data. At the same time, it pre-processes and controls the quality of the collected data to remove noise and outliers, thereby improving the reliability and availability of the data.

[0100] In summary, the multi-source data acquisition module obtains multi-source data such as meteorology, drainage network status and network topology structure comprehensively and in real time through a variety of data acquisition devices and interfaces, providing a solid data foundation for subsequent analysis and decision-making of the urban smart drainage management system.

[0101] In this embodiment, the dynamic 3D modeling module constructs a 3D model of the urban drainage network including network material properties, terrain slope, and node connection relationships based on the principles of fluid mechanics, and dynamically binds real-time sensor data to the 3D model of the urban drainage network through a data fusion algorithm;

[0102] Furthermore, the dynamic 3D modeling module adopts a spatiotemporal data fusion algorithm, and its fusion weight coefficient is calculated in the following way:

[0103] Sensor data contribution part: multiply the real-time reading of each sensor by the geographic influence factor and sum it, then divide it by the maximum reading of the sensor for normalization, and multiply it by the dynamic adjustment parameter α;

[0104] Satellite data contribution: The rainfall forecast value of each satellite meteorological channel is multiplied by the time attenuation coefficient and then summed, divided by the maximum value of the satellite data for normalization, and multiplied by the dynamic adjustment parameter β;

[0105] The final fusion weight coefficient is the weighted sum of the above two parts, where α+β=1 and is updated every 5 minutes;

[0106] The geographical impact factor is calculated based on the node altitude and surrounding building density, with a value range of 0.8-1.2;

[0107] The time attenuation coefficient decays dynamically according to an exponential function, expressed as 0.5 times e to the power of -0.1t;

[0108] The geographic impact factor is calculated using the following rules:

[0109] Altitude influence term: the normalized value of the node altitude relative to the regional extreme value multiplied by 0.2;

[0110] Building density impact term: 1 minus the ratio of the building density around the node to the maximum density of the city, multiplied by 0.8;

[0111] The final geographic impact factor is the sum of the above two items;

[0112] Specifically, the dynamic 3D modeling module is the core component of the urban smart drainage management system based on big data analysis. It is mainly responsible for building a 3D model of the urban drainage network and dynamically binding real-time sensor data to it to achieve accurate simulation and real-time monitoring of the drainage system. The following is a detailed description of the module:

[0113] Overall function overview:

[0114] The dynamic 3D modeling module aims to use the data obtained by the multi-source data acquisition module and combine it with the principles of fluid mechanics to build a 3D model that can reflect the real situation of the urban drainage network. This model not only takes into account the physical structure of the network, but also can be dynamically adjusted according to real-time sensor data, thereby providing accurate basic data for subsequent drainage limit analysis, intelligent early warning and other modules.

[0115] Basic data for modeling:

[0116] The construction of the dynamic 3D modeling module relies on a variety of data provided by the multi-source data acquisition module, including:

[0117] Pipeline material properties: Pipelines of different materials have different water flow resistance, corrosion resistance and other properties, which will affect the flow state of water in the pipe network; for example, the inner wall of a plastic pipe network is relatively smooth and has less water flow resistance; while a metal pipe network may increase water flow resistance due to corrosion;

[0118] Topographic slope: The topographic slope determines the natural flow direction and speed of water flow, and is an important factor affecting the drainage effect. In low-lying areas, water flow tends to accumulate, requiring a more reasonable drainage design.

[0119] Node connection relationship: clarify the connection mode and topological structure between each node in the drainage network, which is crucial for accurately simulating the flow path of water in the network; for example, the connection mode of nodes may include series connection, parallel connection, etc. Different connection modes will result in different distribution and flow of water at the nodes;

[0120] Real-time sensor data: including water level sensor data at each node of the drainage network, drainage flow sensor data, etc.; these data reflect the real-time operation status of the network and are the key to achieving dynamic model updates;

[0121] 3D model construction:

[0122] Based on the principles of fluid mechanics:

[0123] This module builds a three-dimensional model of the urban drainage network based on the principles of fluid mechanics. The principles of fluid mechanics describe the flow of fluids (such as water) in pipes. By applying these principles, a mathematical model that can accurately simulate the flow of water in the drainage network can be established. For example, based on the Bernoulli equation and the continuity equation, the water flow velocity, pressure and other parameters at different locations in the network can be calculated.

[0124] Model building process:

[0125] Geometric modeling: Based on the network topology data, a three-dimensional geometric model of the drainage network is established to determine the shape, size and spatial position of the network;

[0126] Physical property assignment: Assign data such as pipe network material properties and terrain slope to each element in the geometric model so that the model can reflect the real physical characteristics;

[0127] Boundary condition setting: Set the boundary conditions of the model according to the actual situation, such as inlet flow, outlet pressure, etc., to ensure that the model can accurately simulate the actual drainage process;

[0128] Data fusion algorithm:

[0129] In order to enable the 3D model to reflect the operating status of the drainage network in real time, this module uses a spatiotemporal data fusion algorithm to dynamically bind real-time sensor data to the model; the specific formula is as follows: (in, is the fusion weight coefficient, which is used to quantify the contribution of sensor and satellite data; For the The sensors in Real-time reading at the moment, in m³ / s; is the geographical influence factor, with a value range of 0.8-1.2, calculated from the node altitude and the surrounding building density; For the The rainfall intensity forecast value of the satellite meteorological channel, in mm / h; is the time attenuation coefficient, according to Dynamic attenuation; , To dynamically adjust the parameters, And it is updated every 5 minutes; It is the maximum reading of the sensor and is used for normalization processing to ensure that the data is compared under a unified standard; is the maximum value of satellite data, used for normalization);

[0130] Through this algorithm, the data collected by sensors in real time and the satellite meteorological forecast data are fused, and the model is dynamically adjusted according to the fusion weight coefficient, so that the model can reflect the actual operation status of the drainage network in real time;

[0131] Dynamic update and application of models:

[0132] Dynamic update: As real-time sensor data is continuously updated, the model will be adjusted in real time according to the data fusion algorithm to ensure that the model is always consistent with the actual operation status of the drainage network; for example, when the water level sensor data of a node changes, the model will adjust the water flow simulation results of the node and its surrounding areas accordingly;

[0133] Application scenarios: The dynamic 3D model can provide an accurate pipe network model for the drainage limit analysis module to calculate the maximum drainage threshold of each node under extreme rainfall conditions; provide real-time drainage status information for the intelligent early warning module to generate drainage pressure heat maps and trigger early warning mechanisms; and provide data support for the health assessment module and resource scheduling engine to help assess the health of the pipe network and develop reasonable scheduling plans.

[0134] In summary, the dynamic 3D modeling module provides a solid foundation for the efficient operation of the urban smart drainage management system by constructing an accurate 3D model of the urban drainage network and dynamically binding it with real-time sensor data.

[0135] In this embodiment, the drainage limit analysis module uses a pressure-flow coupling algorithm to calculate the maximum drainage threshold of each node under extreme rainfall conditions, and generates a node failure probability distribution diagram based on Monte Carlo simulation;

[0136] Furthermore, the failure probability calculation method of the drainage limit analysis module is:

[0137] Conduct at least 10,000 Monte Carlo simulations, and calculate the actual drainage flow of the node in each simulation;

[0138] The actual discharge is compared to a threshold value obtained by subtracting a correction term for historical bias from the theoretical maximum discharge;

[0139] The proportion of the simulation times when the actual traffic exceeds the threshold to the total times is counted as the node failure probability;

[0140] Correction mechanism: The historical deviation correction term is the historical maximum flow fluctuation value multiplied by the dynamic adjustment factor γ. γ is adaptively adjusted through the hyperbolic tangent function. The adjustment range is affected by the difference between the current flow fluctuation and the standard deviation of historical data.

[0141] The update rule of the dynamic adjustment factor γ is:

[0142] Based on the benchmark value of 0.8;

[0143] The adjustment range is calculated by the difference between the current flow fluctuation and the historical fluctuation through the hyperbolic tangent function;

[0144] Adjust the amplitude coefficient μ to control the range of change, with a value of 0.2-0.6;

[0145] Specifically, the drainage limit analysis module plays a key role in the urban smart drainage management system based on big data analysis. Its main function is to accurately evaluate the carrying capacity of the urban drainage network under extreme rainfall conditions, identify potential risk nodes, and provide an important basis for subsequent early warning and scheduling. The following is a detailed description of the module:

[0146] Overall function overview:

[0147] The drainage limit analysis module uses the pressure-flow coupling algorithm and the improved Monte Carlo simulation algorithm, combined with various data provided by the multi-source data acquisition module, to calculate the maximum drainage threshold of each node in the drainage network under extreme rainfall conditions, and generate a node failure probability distribution map to help the system understand the weak links of the drainage network under extreme conditions;

[0148] Pressure-flow coupling algorithm:

[0149] Algorithm principle:

[0150] The algorithm is based on the principles of fluid mechanics and takes into account the relationship between pressure and flow in the drainage network. In the drainage system, changes in pressure will affect the flow of water, and changes in flow will in turn affect the pressure distribution. By establishing a pressure-flow coupling equation, the pressure and flow values ​​of each node under different working conditions can be accurately calculated.

[0151] Maximum drainage threshold calculation:

[0152] Using the pressure-flow coupling algorithm, combined with the material properties, pipe diameter, slope and other parameters of the pipe network, the maximum drainage flow that each node can withstand under extreme rainfall conditions is calculated, that is, the maximum drainage threshold. This threshold is an important indicator for measuring the drainage capacity of the node. When the actual drainage flow approaches or exceeds the threshold, the node may experience poor drainage or even failure.

[0153] Improved Monte Carlo simulation algorithm:

[0154] Simulation principle:

[0155] Monte Carlo simulation is a method of simulating the behavior of complex systems through random sampling. In drainage limit analysis, the improved Monte Carlo simulation algorithm considers the impact of various uncertain factors on the drainage system through a large number of random simulations, such as the randomness of rainfall intensity and the uncertainty of pipe network parameters.

[0156] Failure probability calculation:

[0157] The failure probability calculation formula is: (in, is the node failure probability, and its value range is ; For the number of simulations, set Second-rate; For the The actual drainage flow of the node in the simulation, in units of ; is the theoretical maximum discharge capacity, calculated based on pipe diameter and slope; is the historical deviation correction factor, according to Adaptive adjustment; is the historical maximum flow fluctuation value, taking the standard deviation of the data of the past 5 years; is an indicator function, which takes 1 when the condition is met, otherwise takes 0);

[0158] Through multiple simulations, the number of times the actual drainage flow exceeds the corrected threshold is counted, thereby calculating the failure probability of the node;

[0159] Historical bias correction factor:

[0160] Historical bias correction factor Adopting adaptive adjustment strategy, the formula is: (in, is the baseline correction value, the default value is 0.8; To adjust the amplitude coefficient, the value range is 0.2-0.6; It is the current monitored flow fluctuation value, in m³ / s; is the standard deviation of historical data, updated quarterly);

[0161] The introduction of this correction factor makes the prediction of failure probability more accurate and can better reflect the actual situation;

[0162] Node failure probability distribution diagram generation:

[0163] Based on the failure probability of each node calculated by the improved Monte Carlo simulation algorithm, a node failure probability distribution diagram is generated; this diagram intuitively shows the failure risk of each node in the drainage network under extreme rainfall conditions, and the darker the color, the higher the failure probability; through the failure probability distribution diagram, system administrators can quickly identify the weak links in the drainage network and take corresponding measures in advance to prevent and deal with them;

[0164] Application value of the module:

[0165] The output results of the drainage limit analysis module provide an important basis for the intelligent early warning module and resource scheduling engine; the intelligent early warning module can set a reasonable early warning threshold based on the node failure probability and the maximum drainage threshold, and issue early warning information in a timely manner when the actual operation data of the node approaches or exceeds the early warning threshold; the resource scheduling engine can reasonably allocate personnel, equipment and materials based on the failure probability distribution map, give priority to the maintenance and processing of nodes with high failure probability, and improve the ability of the urban drainage system to cope with extreme rainfall events;

[0166] In summary, the drainage limit analysis module uses advanced algorithms and simulation technologies to conduct a comprehensive and accurate analysis of the drainage capacity of the urban drainage network under extreme rainfall conditions, providing strong support for the efficient operation of the urban smart drainage management system.

[0167] In this embodiment, the intelligent early warning module generates a heat map of drainage pressure in the next 72 hours by integrating the long short-term memory network LSTM and the ARIMA time series model, combined with satellite meteorological forecast data, and triggers a multi-level early warning mechanism when the real-time data of the node exceeds the preset safety threshold;

[0168] Furthermore, the hybrid prediction model of the intelligent early warning module performs the following operations:

[0169] Prediction formula logic:

[0170] LSTM prediction component: Generates prediction values ​​based on drainage pressure data in a historical time window (24-72 hours);

[0171] ARIMA forecast component: Generates benchmark forecast values ​​through an autoregressive difference moving average model;

[0172] Satellite correction component: the impact of future rainfall intensity predicted by superimposed satellite on drainage pressure;

[0173] Weight distribution rules:

[0174] The weights of LSTM and ARIMA are dynamically calculated by the ratio of their prediction errors and normalized using the sigmoid function;

[0175] The meteorological sensitivity coefficient controls the contribution intensity of the satellite correction term, and its value range is 0.2-0.8;

[0176] Specifically, the intelligent early warning module is a key part of the urban smart drainage management system based on big data analysis. Its main function is to use multi-source data and combine advanced prediction models to accurately predict the future status of the urban drainage system and issue multi-level early warnings in time when abnormal situations occur to ensure the safe and stable operation of the urban drainage system. The following is a specific and detailed description of this module:

[0177] Overall function overview:

[0178] The intelligent early warning module generates a heat map of drainage pressure in the next 72 hours by integrating the long short-term memory network (LSTM) and the ARIMA time series model, combined with satellite meteorological forecast data. When the real-time data of the drainage network node exceeds the preset safety threshold, a multi-level early warning mechanism is triggered to provide timely and effective decision-making support for urban drainage management departments.

[0179] Data source:

[0180] The data of this module mainly comes from the multi-source data acquisition module, including:

[0181] Satellite weather forecast data: Weather information for a period of time in the future provided by meteorological satellites, such as rainfall intensity, rainfall time and spatial distribution, etc. These data are crucial for predicting pressure changes in drainage systems;

[0182] Real-time monitoring data of drainage network: including water level sensor data and drainage flow sensor data of each node of the drainage network, reflecting the current operation status of the drainage system;

[0183] Hybrid prediction model:

[0184] The intelligent early warning module uses a hybrid prediction model to predict drainage pressure. The formula is as follows: (in, for The predicted value of drainage pressure at the moment, in kPa; is the model weight factor, through Dynamic calculation; is the LSTM time window length, set to 24-72 hours; Satellite prediction correction item, reflecting the future rainfall intensity after time; is the meteorological sensitivity coefficient, ranging from 0.2 to 0.8; is the sigmoid function, used to normalize weight distribution);

[0185] LSTM Model:

[0186] Long Short-Term Memory (LSTM) is a special type of recurrent neural network that excels at processing long-term dependencies in sequence data. In this module, the LSTM model is used to learn complex patterns and trends in historical drainage system data to predict future drainage pressure.

[0187] ARIMA Model:

[0188] The Autoregressive Integrated Moving Average (ARIMA) model is a classic time series forecasting model that can capture the linear trend and seasonal changes of data by processing historical data with autoregression and moving average. In the hybrid forecasting model, the ARIMA model is used to provide a relatively stable forecasting basis.

[0189] Satellite prediction corrections:

[0190] Satellite prediction corrections Satellite meteorological forecast data was introduced to consider the impact of future rainfall on drainage pressure; meteorological sensitivity coefficient Used to adjust the weight of satellite data in the forecast, and flexibly adjust according to different meteorological conditions;

[0191] Model weight factors:

[0192] Model weight factor Through dynamic calculation, the contribution of the LSTM model and the ARIMA model in the prediction is adjusted according to their root mean square error (RMSE). This adaptive adjustment mechanism enables the hybrid prediction model to give full play to the advantages of the two models according to different data characteristics and prediction scenarios, thereby improving the accuracy of the prediction.

[0193] Drainage pressure heat map generation:

[0194] Based on the predicted drainage pressure value for the next 72 hours obtained by the hybrid prediction model, a drainage pressure heat map is generated. The heat map displays the drainage pressure of each node in the drainage network at different times with intuitive color distribution. The darker the color, the greater the drainage pressure. Through the heat map, the urban drainage management department can quickly understand the pressure distribution of the drainage system and discover potential risk areas in a timely manner.

[0195] Multi-level early warning mechanism:

[0196] The intelligent early warning module sets a preset safety threshold. When the real-time data of the node exceeds the threshold, a multi-level early warning mechanism is triggered. The early warning level can be divided according to the degree of exceeding the threshold, for example:

[0197] Level 1 warning: slightly exceeding the safety threshold, which may indicate that there are some minor abnormalities in the drainage system, but have not yet caused serious impact on the normal operation of the system; at this time, the system can issue a prompt message to remind managers to pay attention to the operation of related nodes;

[0198] Level 2 warning: The degree of exceeding the safety threshold is large, indicating that the drainage system may face certain risks and needs to take corresponding measures to deal with it; the system can automatically notify relevant personnel to conduct on-site inspections and provide preliminary treatment suggestions;

[0199] Level 3 warning: If the safety threshold is seriously exceeded, it means that the drainage system may have a fault or face a major risk, and the emergency plan needs to be activated immediately; the system can quickly notify the emergency rescue team and provide detailed fault information and rescue plans;

[0200] Application value of the module:

[0201] The intelligent early warning module helps urban drainage management departments to detect potential problems in the drainage system in advance through accurate prediction and timely warning, take effective measures to prevent and deal with them, and reduce the occurrence of disasters such as urban waterlogging; at the same time, the module's multi-level early warning mechanism can provide corresponding response strategies according to different risk levels, thereby improving the emergency response capability and management efficiency of the urban drainage system.

[0202] In this embodiment, the health assessment module uses the random forest algorithm to analyze the correlation between the node's historical fault data and the real-time performance indicators, generates a node health index including the pipe wall corrosion rate and the sediment accumulation coefficient, and outputs the maintenance priority ranking;

[0203] Furthermore, the health index calculation rule of the health assessment module is:

[0204] Calculate the ratio of the current value to the maximum allowable value for indicators such as corrosion depth and deposition thickness;

[0205] A dynamic weight is assigned to each indicator, which is calculated by the entropy weight method and the sum is 1;

[0206] Multiply the score of each indicator by the exponential decay factor. The decay coefficient λ reflects the aging rate of the pipeline network. The value range of λ is 0.05-0.2 / year.

[0207] The final health index is the normalized value of the weighted sum of all indicator scores;

[0208] Specifically, the health assessment module plays an important role in the urban smart drainage management system based on big data analysis. It mainly uses the random forest algorithm to analyze the node's historical fault data and real-time performance indicators, and then generates a node health index and gives a maintenance priority ranking, providing a key basis for the maintenance and management of the urban drainage network. The following is a detailed and specific description of the module:

[0209] Overall function overview:

[0210] The core task of the health assessment module is to comprehensively assess the health status of each node in the drainage network. With the help of in-depth mining and analysis of historical fault data and real-time performance indicators, the module can accurately identify possible problems and potential risks at the nodes, providing scientific and reasonable decision-making support for subsequent maintenance work.

[0211] Data source:

[0212] The data of this module mainly comes from the multi-source data acquisition module, which specifically covers the following categories:

[0213] Node historical fault data: Contains information such as the time, type, cause, and maintenance record of past node failures. This data reflects the stability and reliability of the node during its historical operation and can provide important clues for analyzing the node's failure modes and patterns.

[0214] Real-time performance indicators: including real-time monitoring data such as pipe wall corrosion rate, sediment accumulation coefficient, water level, flow rate, pressure, etc.; these indicators directly reflect the current operating status and physical characteristics of the node and are an important basis for evaluating the health status of the node;

[0215] Multi-feature fusion evaluation algorithm:

[0216] The health assessment module uses a multi-feature fusion assessment algorithm to calculate the health index of the node. The specific formula is as follows: (in, is the health index, standardized to 0-100 points; For the The dynamic weight of each indicator is calculated by the entropy weight method and meets the ; Indicates the currently detected physical quantities, including corrosion depth, deposition thickness and other physical quantities; The maximum value allowed as specified by industry standards; is the aging attenuation coefficient, ranging from 0.05 to 0.2 per year; The current timestamp, accurate to the day; Installation time for the pipe network; represents the number of metrics used to evaluate the health of a node);

[0217] Dynamic weight of indicators: is calculated by the entropy weight method. The entropy weight method is an objective weighting method, which determines the weight according to the discrete degree of the indicator data. The greater the discrete degree of the indicator, the higher its weight. This can avoid the influence of subjective factors and make the weight distribution more reasonable.

[0218] Current test values ​​and industry standards: Indicates the currently detected physical quantities, such as the corrosion depth of the pipe wall, the thickness of the deposits, etc. is the maximum value allowed for these physical quantities as specified by industry standards; by calculating , we can get the degree of deviation of each indicator relative to the standard value. The greater the deviation, the greater the impact of the indicator on the health status of the node;

[0219] Aging decay effects: It reflects the aging and attenuation effect of the pipe network as the service time increases; is the aging attenuation coefficient, which reflects the aging rate of the pipeline network; as time goes by, The value of increases, The value of decreases, which means that the health of the node will gradually decline due to aging;

[0220] Random Forest Algorithm Analysis:

[0221] This module uses the random forest algorithm to analyze the correlation between node historical fault data and real-time performance indicators. Random forest is an ensemble learning method that consists of multiple decision trees. By learning from a large amount of historical data, the random forest algorithm can discover the complex relationship between different indicators and their correlation patterns with node failures.

[0222] During the training process, the random forest algorithm will randomly select a portion of data and features to build a decision tree, and then combine the results of multiple decision trees; this can improve the accuracy and stability of the model and reduce the risk of overfitting; by analyzing the output results of the random forest algorithm, it is possible to determine which indicators have a greater impact on the health status of the node, providing a more scientific basis for the calculation of the health index;

[0223] Health Index and Maintenance Priority Ranking:

[0224] Health Index Calculation:

[0225] According to the multi-feature fusion evaluation algorithm, the various indicators of each node are comprehensively calculated to obtain the node health index The health index ranges from 0 to 100. A higher score indicates a better health status of the node. A lower score indicates that the node has more problems and needs timely maintenance.

[0226] Maintenance Prioritization:

[0227] Based on the calculated health index, all nodes are prioritized for maintenance; nodes with lower health indexes are ranked first and given priority for maintenance; this ensures that limited maintenance resources can be allocated to the nodes that need them most, improving maintenance efficiency and reducing the risk of failure in the drainage network;

[0228] Application value of the module:

[0229] The health assessment module provides a clear basis for maintenance decision-making for urban drainage management departments by scientifically and accurately evaluating the health status of drainage network nodes. It can help management departments discover potential problems in advance, arrange maintenance plans reasonably, extend the service life of the network, and ensure the stable operation of the urban drainage system. At the same time, by sorting maintenance priorities, it optimizes the allocation of maintenance resources and improves the efficiency and benefits of maintenance work.

[0230] In this embodiment, the resource scheduling engine optimizes maintenance path planning based on genetic algorithms, generates dynamic scheduling plans based on the warning level, health index, and emergency resource distribution, and realizes the real-time optimal configuration of personnel, equipment, and materials;

[0231] Furthermore, the optimization objective function of the resource scheduling engine is defined as:

[0232] Total path time: the sum of the path distances of each rescue point divided by the dynamic driving speed;

[0233] Time breach cost: impose a penalty on the response time that exceeds the time limit of the plan, and the penalty coefficient η is positively correlated with the warning level;

[0234] Total resource constraints: The available resources must meet the demand threshold;

[0235] Decision variable constraints: Resource allocation uses 0-1 integer programming;

[0236] Further, the optimization based on genetic algorithm performs the following steps:

[0237] The non-dominated sorting genetic algorithm NSGA-II is used to generate the Pareto optimal solution set;

[0238] Define the dual objective vector: total path time and time breach cost;

[0239] Select the optimal solution based on the real-time weather warning level, and prioritize minimizing the time default cost when a red warning occurs;

[0240] Specifically, the resource scheduling engine is one of the core modules in the urban smart drainage management system based on big data analysis. Its main function is to make real-time and optimal configuration of emergency resources such as personnel, equipment, and materials based on the information provided by each module in the system, so as to efficiently deal with various problems arising in the urban drainage system and ensure the stable operation of the urban drainage system. The following is a detailed and specific description of this module:

[0241] Overall function overview:

[0242] The resource scheduling engine comprehensively considers multiple factors such as warning level, node health index and emergency resource distribution, uses genetic algorithms to optimize the maintenance path planning, generates dynamic scheduling plans, and achieves the optimal allocation of emergency resources in time and space, thereby improving the ability of the urban drainage system to respond to emergencies;

[0243] Data input:

[0244] The operation of the resource scheduling engine depends on the data provided by multiple modules, as follows:

[0245] Warning level: generated by the intelligent warning module based on drainage pressure prediction and real-time monitoring data, reflecting the risk level faced by each node in the drainage system; the warning level is divided into different levels, such as level 1 warning, level 2 warning, level 3 warning, etc. The higher the level, the more serious the risk;

[0246] Node health index: calculated by the health assessment module, reflecting the health status of each node in the drainage network; the lower the health index, the more serious the problem of the node, and the more priority maintenance is needed;

[0247] Emergency resource distribution: including the location and quantity of emergency resources such as personnel, equipment, and materials in the city; these resources are the basis for responding to drainage system failures and disasters;

[0248] Multi-objective optimization algorithm:

[0249] The resource scheduling engine uses a multi-objective optimization algorithm to determine the optimal resource scheduling solution. The objective function is as follows: (in, For the The path distance to each rescue point, in km; It is the dynamic driving speed, obtained through the real-time traffic API, in km / h; is the time penalty factor, which is set to 10-50 and is positively correlated with the warning level; is the actual response time in minutes; Set time limits for emergency plans, divided according to warning levels; represents the number of rescue points);

[0250] Path time optimization: Indicates the total time required for rescue personnel or equipment to reach the fault node from the resource point; obtain dynamic driving speed through the real-time traffic API , the path time can be calculated more accurately, so as to select the optimal driving path and reduce the response time;

[0251] Time penalty: It is a time penalty item, which is used to punish the situation where the actual response time exceeds the time limit specified in the plan; is the time penalty factor, which is positively correlated with the warning level. The higher the warning level, The larger the value, the more severe the penalty for timeouts; this helps ensure timely response and reduce losses in high-risk situations;

[0252] Constraints:

[0253] When scheduling resources, the following constraints must be met:

[0254] Resource requirement constraints: ,in, is the available resource amount, is the demand threshold; this ensures that the scheduled resources can meet the maintenance and processing needs of the faulty nodes;

[0255] Resource allocation constraints: , indicating whether to start from the resource point To Node Allocate resources; Indicates the allocation of resources. Indicates that resources are not allocated;

[0256] Genetic algorithm optimization:

[0257] In order to solve the multi-objective optimization problem, the resource scheduling engine uses genetic algorithms to optimize the maintenance path planning. Genetic algorithms are an optimization algorithm that simulates natural selection and genetic mechanisms, and finds the optimal solution through continuous iterative evolution.

[0258] coding:

[0259] The resource scheduling scheme is encoded as a chromosome, each chromosome represents a possible scheduling scheme; the genes in the chromosome represent the allocation of resources and path selection;

[0260] Fitness function:

[0261] The fitness value of each chromosome is calculated according to the objective function. The higher the fitness value, the better the scheduling scheme.

[0262] Selection, crossover, and mutation:

[0263] Through genetic operations such as selection, crossover and mutation, new chromosomes are continuously generated, gradually approaching the optimal solution; the selection operation selects excellent chromosomes according to the fitness value, the crossover operation exchanges the genes of excellent chromosomes, and the mutation operation randomly mutates the genes of chromosomes to increase the diversity of the population;

[0264] Pareto optimal solution screening mechanism:

[0265] In order to achieve a balance between multiple optimization objectives, the resource scheduling engine introduces a Pareto optimal solution screening mechanism; the optimization objective function is defined as: ,in, It is Path distance to each rescue point (in km), is the dynamic driving speed;

[0266] The Pareto frontier is solved by the non-dominated sorting genetic algorithm (NSGA-II) to obtain a set of non-dominated solutions; the final solution is selected by the decision maker based on the real-time meteorological warning level and actual conditions to achieve a balanced decision between efficiency and safety;

[0267] Dynamic scheduling scheme generation:

[0268] The resource scheduling engine generates a dynamic scheduling plan based on the calculation results of the multi-objective optimization algorithm and the genetic algorithm, combined with the changes in real-time data. The plan specifies the type and quantity of resources required for each faulty node, as well as the resource allocation path and time schedule. At the same time, as the warning level, node health index, and emergency resource distribution change, the scheduling plan will be updated in real time to ensure the optimal configuration of resources.

[0269] Application value of the module:

[0270] The resource scheduling engine improves the emergency response capability and maintenance efficiency of the urban drainage system through scientific and reasonable resource allocation and path planning; it can allocate emergency resources to where they are most needed in the shortest time, reduce the impact of disasters such as urban waterlogging, and protect the safety of life and property of urban residents and the normal operation of the city; at the same time, the real-time update mechanism of the dynamic scheduling plan ensures the flexibility and adaptability of the system, and can cope with various complex and changing situations.

[0271] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An urban smart drainage management system based on big data analysis, characterized in that: include: Multi-source data acquisition module, used to obtain historical climate change data, high-precision historical rainfall curves, satellite meteorological monitoring data, water level sensor data at each node of the drainage network, drainage flow sensor data and network topology data in real time; The dynamic 3D modeling module builds a 3D model of the urban drainage network that includes network material properties, terrain slope, and node connection relationships based on the principles of fluid mechanics, and dynamically binds real-time sensor data to the 3D model of the urban drainage network through a data fusion algorithm; The drainage limit analysis module uses the pressure-flow coupling algorithm to calculate the maximum drainage threshold of each node under extreme rainfall conditions, and generates a node failure probability distribution diagram based on Monte Carlo simulation; The intelligent early warning module generates a heat map of drainage pressure in the next 72 hours by integrating the long short-term memory network LSTM and the ARIMA time series model, combined with satellite meteorological forecast data. When the real-time data of the node exceeds the preset safety threshold, a multi-level early warning mechanism is triggered; The health assessment module uses the random forest algorithm to analyze the correlation between the node's historical fault data and real-time performance indicators, generates a node health index including the pipe wall corrosion rate and sediment accumulation coefficient, and outputs the maintenance priority ranking; The resource scheduling engine optimizes maintenance path planning based on genetic algorithms, generates dynamic scheduling plans based on comprehensive warning levels, health indexes, and emergency resource distribution, and achieves real-time optimal configuration of personnel, equipment, and materials.

2. The urban smart drainage management system based on big data analysis according to claim 1 is characterized in that: The multi-source data acquisition module comprises: Multi-parameter sensor arrays deployed at key nodes of the drainage network, including ultrasonic flow meters, laser turbidity meters, and pressure transmitters; Meteorological satellite data interface, supporting the reception of high-resolution synthetic aperture radar SAR rainfall monitoring data; The municipal geographic information system GIS data fusion unit realizes the three-dimensional coordinate calibration of the pipe network topology structure.

3. The urban smart drainage management system based on big data analysis according to claim 2 is characterized in that: The dynamic 3D modeling module adopts a spatiotemporal data fusion algorithm, and its fusion weight coefficient is calculated in the following way: Sensor data contribution part: multiply the real-time reading of each sensor by the geographic influence factor and sum it, then divide it by the maximum reading of the sensor for normalization, and multiply it by the dynamic adjustment parameter α; Satellite data contribution: The rainfall forecast value of each satellite meteorological channel is multiplied by the time attenuation coefficient and then summed, divided by the maximum value of the satellite data for normalization, and multiplied by the dynamic adjustment parameter β; The final fusion weight coefficient is the weighted sum of the above two parts, where α+β=1 and is updated every 5 minutes; The geographical impact factor is calculated based on the node altitude and surrounding building density, with a value range of 0.8-1.2; The time attenuation coefficient decays dynamically according to an exponential function, expressed as 0.5 multiplied by e to the power of -0.1t.

4. The urban smart drainage management system based on big data analysis according to claim 3 is characterized in that: The failure probability calculation method of the drainage limit analysis module is: Conduct at least 10,000 Monte Carlo simulations, and calculate the actual drainage flow of the node in each simulation; The actual discharge is compared to a threshold value obtained by subtracting a correction term for historical bias from the theoretical maximum discharge; The proportion of the simulation times when the actual traffic exceeds the threshold to the total times is counted as the node failure probability; Correction mechanism: The historical deviation correction term is the multiplication of the historical maximum flow fluctuation value and the dynamic adjustment factor γ. γ is adaptively adjusted through the hyperbolic tangent function, and the adjustment range is affected by the difference between the current flow fluctuation and the standard deviation of the historical data.

5. The urban smart drainage management system based on big data analysis according to claim 4 is characterized in that: The hybrid prediction model of the intelligent early warning module performs the following operations: Prediction formula logic: LSTM prediction component: Generates prediction values ​​based on drainage pressure data in a historical time window (24-72 hours); ARIMA forecast component: Generates benchmark forecast values ​​through an autoregressive difference moving average model; Satellite correction component: the impact of future rainfall intensity predicted by superimposed satellite on drainage pressure; Weight distribution rules: The weights of LSTM and ARIMA are dynamically calculated by the ratio of their prediction errors and normalized using the sigmoid function; The meteorological sensitivity coefficient controls the contribution intensity of the satellite correction term and has a value range of 0.2-0.

8.

6. The urban smart drainage management system based on big data analysis according to claim 5 is characterized in that: The health index calculation rule of the health assessment module is: Calculate the ratio of the current value to the maximum allowable value for indicators such as corrosion depth and deposition thickness; A dynamic weight is assigned to each indicator, which is calculated by the entropy weight method and the sum is 1; Multiply the score of each indicator by the exponential decay factor. The decay coefficient λ reflects the aging rate of the pipeline network. The value range of λ is 0.05-0.2 / year. The final health index is the standardized value of the weighted sum of all indicator scores.

7. The urban smart drainage management system based on big data analysis according to claim 6 is characterized in that: The optimization objective function of the resource scheduling engine is defined as: Total path time: the sum of the path distances of each rescue point divided by the dynamic driving speed; Time breach cost: impose a penalty on the response time that exceeds the time limit of the plan, and the penalty coefficient η is positively correlated with the warning level; Total resource constraints: The available resources must meet the demand threshold; Decision variable constraints: Resource allocation adopts 0-1 integer programming.

8. The urban smart drainage management system based on big data analysis according to claim 7 is characterized in that: The geographic impact factor is calculated by the following rules: Altitude influence term: the normalized value of the node altitude relative to the regional extreme value multiplied by 0.2; Building density impact term: 1 minus the ratio of the building density around the node to the maximum density of the city, multiplied by 0.8; The final geographic impact factor is the sum of the above two items.

9. The urban smart drainage management system based on big data analysis according to claim 8 is characterized in that: The update rule of the dynamic adjustment factor γ is: Based on the benchmark value of 0.8; The adjustment range is calculated by the difference between the current flow fluctuation and the historical fluctuation through the hyperbolic tangent function; Adjust the amplitude coefficient μ to control the range of change, with a value of 0.2-0.

6.

10. The urban smart drainage management system based on big data analysis according to claim 9 is characterized in that: The genetic algorithm based optimization performs the following steps: The non-dominated sorting genetic algorithm NSGA-II is used to generate the Pareto optimal solution set; Define the dual objective vector: total path time and time breach cost; The optimal solution is selected according to the real-time weather warning level, and the time default cost is minimized first when the red warning occurs.

Images